v0.12.0

feat(ollama): add support for base64-encoded images in chat messages and forward them to the Ollama provider
v0.11.1
2026-01-20 01:10:27 +00:00 · 2026-01-20 01:10:27 +00:00 · 2026-01-20 00:37:59 +00:00 · 2026-01-20 00:37:49 +00:00 · 2026-01-20 00:12:21 +00:00 · 2026-01-20 00:12:21 +00:00
36 changed files with 3735 additions and 3397 deletions
--- a/.gitignore
+++ b/.gitignore
@@ -3,7 +3,6 @@
 # artifacts
 coverage/
 public/
 pages/
 # installs
 node_modules/
@@ -17,4 +16,8 @@ node_modules/
 dist/
 dist_*/
-# custom
+# AI
 .claude/
 .serena/
 #------# custom
--- a/changelog.md
+++ b/changelog.md
@@ -1,6 +1,58 @@
 # Changelog
 ## 2026-01-20 - 0.12.0 - feat(ollama)
 add support for base64-encoded images in chat messages and forward them to the Ollama provider
 - Add optional images?: string[] to ChatMessage and ChatOptions interfaces (multimodal/vision support)
 - Propagate images from messageHistory and ChatOptions to the Ollama API payload in chat, chatStreaming, and streaming handlers
 - Changes are non-breaking: images are optional and existing behavior is preserved when absent
 ## 2026-01-20 - 0.11.0 - feat(ollama)
 support defaultOptions and defaultTimeout for ollama provider
 - Added ollama.defaultOptions object with fields: num_ctx, temperature, top_k, top_p, repeat_penalty, num_predict, stop, seed
 - Added ollama.defaultTimeout option
 - Pass defaultOptions and defaultTimeout into OllamaProvider constructor when initializing the provider
 - Non-breaking change: existing behavior preserved if new fields are undefined
 ## 2026-01-20 - 0.10.1 - fix()
 no changes detected — no release necessary
 - No files changed in the provided diff; there are no code, documentation, or configuration modifications to release.
 ## 2026-01-18 - 0.10.0 - feat(mistral)
 add Mistral provider with native PDF OCR and chat integration
 - Adds dependency @mistralai/mistralai
 - Implements ts/provider.mistral.ts providing chat() and document() (OCR) functionality
 - Registers and exposes MistralProvider in SmartAi (options, lifecycle, conversation routing)
 - Adds unit/integration tests: test.chat.mistral.ts and test.document.mistral.ts
 - Updates readme.hints.md with Mistral usage, configuration and notes
 ## 2026-01-18 - 0.9.0 - feat(providers)
 Add Anthropic extended thinking and adapt providers to new streaming/file APIs; bump dependencies and update docs, tests and configuration
 - Add IAnthropicProviderOptions.extendedThinking with thinking modes (quick/normal/deep/off) and getThinkingConfig mapping budgets; apply thinking to Anthropic requests and omit temperature when thinking is enabled.
 - Update Anthropic research flow to include thinking configuration and conditionally set temperature.
 - OpenAI image editing: use openai.toFile to convert image/mask Buffers to uploadable files (image/png) before sending.
 - ElevenLabs streaming: switch from response.streamNode() to response.stream() and convert web stream to Node stream using Readable.fromWeb().
 - Upgrade dependencies and dev tools: @anthropic-ai/sdk ^0.71.2, @push.rocks/smartrequest ^5.0.1, @git.zone/tsbuild and related @git.zone packages, and other bumps in package.json.
 - Tests and test imports updated to use @git.zone/tstest/tapbundle; many test files adjusted accordingly.
 - Docs and hints updated: README and readme.hints.md include extended thinking docs, examples, formatting fixes, security/issue reporting guidance, and trademark/license clarifications.
 - Project config tweaks: package build script changed, tsconfig baseUrl/paths added, npmextra.json reorganized (release registries added), .gitignore updated to ignore .claude/.serena local tooling files.
 ## 2025-10-30 - 0.8.0 - feat(provider.anthropic)
 Add extended thinking modes to AnthropicProvider and apply thinking budgets to API calls
 - Introduce IAnthropicProviderOptions.extendedThinking to configure thinking modes: 'quick' | 'normal' | 'deep' | 'off'.
 - Add getThinkingConfig() helper mapping modes to token budgets (quick=2048, normal=8000, deep=16000, off=0).
 - Apply thinking configuration to Anthropic API calls (chat, chatStream, vision, document, research) and increase max_tokens where appropriate (up to 20000).
 - Add comprehensive tests (test/test.thinking.anthropic.ts) and update readme.hints.md with usage examples and recommendations.
 - Add .claude/settings.local.json for local assistant permissions used in development/testing.
 ## 2025-10-10 - 0.7.7 - fix(MultiModalModel)
 Lazy-load SmartPdf and guard document processing across providers; ensure SmartPdf is initialized only when needed
 - Make SmartPdf lazy-loaded: smartpdfInstance is now nullable and no longer started automatically in start()
@@ -11,6 +63,7 @@ Lazy-load SmartPdf and guard document processing across providers; ensure SmartP
 - Add local Claude permissions file (.claude/settings.local.json) for tooling/configuration
 ## 2025-10-09 - 0.7.6 - fix(provider.elevenlabs)
 Provide default ElevenLabs TTS voice fallback and add local tool/project configs
 - ElevenLabsProvider: fallback to Samara voice id ('19STyYD15bswVz51nqLf') when no voiceId or defaultVoiceId is provided — avoids throwing an error on TTS calls.
@@ -19,18 +72,21 @@ Provide default ElevenLabs TTS voice fallback and add local tool/project configs
 - Add .serena/project.yml and .serena/.gitignore to include Serena project configuration and ignore cache.
 ## 2025-10-08 - 0.7.5 - fix(provider.elevenlabs)
 Update ElevenLabs default TTS model to eleven_v3 and add local Claude permissions file
 - Changed default ElevenLabs modelId from 'eleven_multilingual_v2' to 'eleven_v3' in ts/provider.elevenlabs.ts to use the newer/default TTS model.
 - Added .claude/settings.local.json with a permissions allow-list for local Claude tooling and CI tasks.
 ## 2025-10-03 - 0.7.4 - fix(provider.anthropic)
 Use image/png for embedded PDF images in Anthropic provider and add local Claude settings for development permissions
 - AnthropicProvider: change media_type from 'image/jpeg' to 'image/png' when embedding images extracted from PDFs to ensure correct format in Anthropic requests.
 - Add .claude/settings.local.json with development/testing permissions for local Claude usage (shell commands, webfetch, websearch, test/run tasks).
 ## 2025-10-03 - 0.7.3 - fix(tests)
 Add extensive provider/feature tests and local Claude CI permissions
 - Add many focused test files covering providers and features: OpenAI, Anthropic, Perplexity, Groq, Ollama, Exo, XAI (chat, audio, vision, document, research, image generation, stubs, interfaces, basic)
@@ -39,12 +95,14 @@ Add extensive provider/feature tests and local Claude CI permissions
 - No changes to library runtime code — this change adds tests and CI/local agent configuration only
 ## 2025-10-03 - 0.7.2 - fix(anthropic)
 Update Anthropic provider branding to Claude Sonnet 4.5 and add local Claude permissions
 - Docs: Replace 'Claude 3 Opus' with 'Claude Sonnet 4.5' in README provider capabilities matrix.
 - Config: Add .claude/settings.local.json to define local Claude permissions for tests and development commands.
 ## 2025-10-03 - 0.7.1 - fix(docs)
 Add README image generation docs and .claude local settings
 - Add .claude/settings.local.json with permission allow-list for local assistant tooling and web search
@@ -53,6 +111,7 @@ Add README image generation docs and .claude local settings
 - Mark image generation support as implemented in the roadmap and remove duplicate entry
 ## 2025-10-03 - 0.7.0 - feat(providers)
 Add research API and image generation/editing support; extend providers and tests
 - Introduce ResearchOptions and ResearchResponse to the MultiModalModel interface and implement research() where supported
@@ -66,14 +125,16 @@ Add research API and image generation/editing support; extend providers and test
 - Add local Claude agent permissions file (.claude/settings.local.json) and various provider type/import updates
 ## 2025-09-28 - 0.6.1 - fix(provider.anthropic)
 Fix Anthropic research tool identifier and add tests + local Claude permissions
 - Replace Anthropic research tool type from 'computer_20241022' to 'web_search_20250305' to match the expected web-search tool schema.
- Add comprehensive test suites and fixtures for providers and research features (new/updated tests under test/ including anthropic, openai, research.* and stubs).
+- Add comprehensive test suites and fixtures for providers and research features (new/updated tests under test/ including anthropic, openai, research.\* and stubs).
 - Fix test usage of XAI provider class name (use XAIProvider) and adjust basic provider test expectations (provider instantiation moved to start()).
 - Add .claude/settings.local.json with local Claude permissions to allow common CI/dev commands and web search during testing.
 ## 2025-09-28 - 0.6.0 - feat(research)
 Introduce research API with provider implementations, docs and tests
 - Add ResearchOptions and ResearchResponse interfaces and a new abstract research() method to MultiModalModel
@@ -88,6 +149,7 @@ Introduce research API with provider implementations, docs and tests
 - Add .claude/settings.local.json (local agent permissions for CI/dev tasks)
 ## 2025-08-12 - 0.5.11 - fix(openaiProvider)
 Update default chat model to gpt-5-mini and bump dependency versions
 - Changed default chat model in OpenAiProvider from 'o3-mini' and 'o4-mini' to 'gpt-5-mini'
@@ -96,6 +158,7 @@ Update default chat model to gpt-5-mini and bump dependency versions
 - Added new local Claude settings configuration (.claude/settings.local.json)
 ## 2025-08-03 - 0.5.10 - fix(dependencies)
 Update SmartPdf to v4.1.1 for enhanced PDF processing capabilities
 - Updated @push.rocks/smartpdf from ^3.3.0 to ^4.1.1
@@ -103,12 +166,14 @@ Update SmartPdf to v4.1.1 for enhanced PDF processing capabilities
 - Dependency updates for better performance and compatibility
 ## 2025-08-01 - 0.5.9 - fix(documentation)
 Remove contribution section from readme
 - Removed the contribution section from readme.md as requested
 - Kept the roadmap section for future development plans
 ## 2025-08-01 - 0.5.8 - fix(core)
 Fix SmartPdf lifecycle management and update dependencies
 - Moved SmartPdf instance management to the MultiModalModel base class for better resource sharing
@@ -118,12 +183,14 @@ Fix SmartPdf lifecycle management and update dependencies
 - Enhanced readme with professional documentation and feature matrix
 ## 2025-07-26 - 0.5.7 - fix(provider.openai)
 Fix stream type mismatch in audio method
 - Fixed type error where OpenAI SDK returns a web ReadableStream but the audio method needs to return a Node.js ReadableStream
 - Added conversion using Node.js's built-in Readable.fromWeb() method
 ## 2025-07-25 - 0.5.5 - feat(documentation)
 Comprehensive documentation enhancement and test improvements
 - Completely rewrote readme.md with detailed provider comparisons, advanced usage examples, and performance tips
@@ -132,6 +199,7 @@ Comprehensive documentation enhancement and test improvements
 - Added verbose flag to test script for better debugging
 ## 2025-05-13 - 0.5.4 - fix(provider.openai)
 Update dependency versions, clean test imports, and adjust default OpenAI model configurations
 - Bump dependency versions in package.json (@git.zone/tsbuild, @push.rocks/tapbundle, openai, etc.)
@@ -139,17 +207,20 @@ Update dependency versions, clean test imports, and adjust default OpenAI model
 - Remove unused 'expectAsync' import from test file
 ## 2025-04-03 - 0.5.3 - fix(package.json)
 Add explicit packageManager field to package.json
 - Include the packageManager property to specify the pnpm version and checksum.
 - Align package metadata with current standards.
 ## 2025-04-03 - 0.5.2 - fix(readme)
 Remove redundant conclusion section from README to streamline documentation.
 - Eliminated the conclusion block describing SmartAi's capabilities and documentation pointers.
 ## 2025-02-25 - 0.5.1 - fix(OpenAiProvider)
 Corrected audio model ID in OpenAiProvider
 - Fixed audio model identifier from 'o3-mini' to 'tts-1-hd' in the OpenAiProvider's audio method.
@@ -157,6 +228,7 @@ Corrected audio model ID in OpenAiProvider
 - Corrected spelling errors in test documentation and comments.
 ## 2025-02-25 - 0.5.0 - feat(documentation and configuration)
 Enhanced package and README documentation
 - Expanded the package description to better reflect the library's capabilities.
@@ -164,6 +236,7 @@ Enhanced package and README documentation
 - Provided error handling strategies and advanced streaming customization examples.
 ## 2025-02-25 - 0.4.2 - fix(core)
 Fix OpenAI chat streaming and PDF document processing logic.
 - Updated OpenAI chat streaming to handle new async iterable format.
@@ -171,6 +244,7 @@ Fix OpenAI chat streaming and PDF document processing logic.
 - Removed unsupported temperature options from OpenAI requests.
 ## 2025-02-25 - 0.4.1 - fix(provider)
 Fix provider modules for consistency
 - Updated TypeScript interfaces and options in provider modules for better type safety.
@@ -178,6 +252,7 @@ Fix provider modules for consistency
 - Added optional model options to OpenAI provider for custom model usage.
 ## 2025-02-08 - 0.4.0 - feat(core)
 Added support for Exo AI provider
 - Introduced ExoProvider with chat functionalities.
@@ -185,18 +260,21 @@ Added support for Exo AI provider
 - Extended Conversation class to support ExoProvider.
 ## 2025-02-05 - 0.3.3 - fix(documentation)
 Update readme with detailed license and legal information.
 - Added explicit section on License and Legal Information in the README.
 - Clarified the use of trademarks and company information.
 ## 2025-02-05 - 0.3.2 - fix(documentation)
 Remove redundant badges from readme
 - Removed Build Status badge from the readme file.
 - Removed License badge from the readme file.
 ## 2025-02-05 - 0.3.1 - fix(documentation)
 Updated README structure and added detailed usage examples
 - Introduced a Table of Contents
@@ -205,6 +283,7 @@ Updated README structure and added detailed usage examples
 - Clarified the development setup with instructions for running tests and building the project
 ## 2025-02-05 - 0.3.0 - feat(integration-xai)
 Add support for X.AI provider with chat and document processing capabilities.
 - Introduced XAIProvider class for integrating X.AI features.
@@ -212,6 +291,7 @@ Add support for X.AI provider with chat and document processing capabilities.
 - Enabled document processing capabilities with PDF conversion in X.AI.
 ## 2025-02-03 - 0.2.0 - feat(provider.anthropic)
 Add support for vision and document processing in Anthropic provider
 - Implemented vision tasks for Anthropic provider using Claude-3-opus-20240229 model.
@@ -219,6 +299,7 @@ Add support for vision and document processing in Anthropic provider
 - Updated documentation to reflect the new capabilities of the Anthropic provider.
 ## 2025-02-03 - 0.1.0 - feat(providers)
 Add vision and document processing capabilities to providers
 - OpenAI and Ollama providers now support vision tasks using GPT-4 Vision and Llava models respectively.
@@ -227,6 +308,7 @@ Add vision and document processing capabilities to providers
 - Updated the readme file with examples for vision and document processing.
 ## 2025-02-03 - 0.0.19 - fix(core)
 Enhanced chat streaming and error handling across providers
 - Refactored chatStream method to properly handle input streams and processes in Perplexity, OpenAI, Ollama, and Anthropic providers.
@@ -235,6 +317,7 @@ Enhanced chat streaming and error handling across providers
 - Adjusted the test logic in test/test.ts for the new classification response requirement.
 ## 2024-09-19 - 0.0.18 - fix(dependencies)
 Update dependencies to the latest versions.
 - Updated @git.zone/tsbuild from ^2.1.76 to ^2.1.84
@@ -248,46 +331,53 @@ Update dependencies to the latest versions.
 - Updated openai from ^4.47.1 to ^4.62.1
 ## 2024-05-29 - 0.0.17 - Documentation
 Updated project description.
 - Improved project description for clarity and details.
 ## 2024-05-17 - 0.0.16 to 0.0.15 - Core
 Fixes and updates.
 - Various core updates and fixes for stability improvements.
 ## 2024-04-29 - 0.0.14 to 0.0.13 - Core
 Fixes and updates.
 - Multiple core updates and fixes for enhanced functionality.
 ## 2024-04-29 - 0.0.12 - Core
 Fixes and updates.
 - Core update and bug fixes.
 ## 2024-04-29 - 0.0.11 - Provider
 Fix integration for anthropic provider.
 - Correction in the integration process with anthropic provider for better compatibility.
 ## 2024-04-27 - 0.0.10 to 0.0.9 - Core
 Fixes and updates.
 - Updates and fixes to core components.
 - Updated tsconfig for improved TypeScript configuration.
 ## 2024-04-01 - 0.0.8 to 0.0.7 - Core and npmextra
 Core updates and npmextra configuration.
 - Core fixes and updates.
 - Updates to npmextra.json for githost configuration.
 ## 2024-03-31 - 0.0.6 to 0.0.2 - Core
 Initial core updates and fixes.
 - Multiple updates and fixes to core following initial versions.
 This summarizes the relevant updates and changes based on the provided commit messages. The changelog excludes commits that are version tags without meaningful content or repeated entries.
--- a/npmextra.json
+++ b/npmextra.json
@@ -1,5 +1,5 @@
 {
-  "gitzone": {
+  "@git.zone/cli": {
    "projectType": "npm",
    "module": {
      "githost": "code.foss.global",
@@ -33,13 +33,19 @@
        "AI toolkit",
        "provider switching"
      ]
    },
    "release": {
      "accessLevel": "public",
      "registries": [
        "https://verdaccio.lossless.digital",
        "https://registry.npmjs.org"
      ]
    }
  },
-  "npmci": {
+  "@git.zone/tsdoc": {
    "npmGlobalTools": [],
    "npmAccessLevel": "public"
  },
  "tsdoc": {
    "legal": "\n## License and Legal Information\n\nThis repository contains open-source code that is licensed under the MIT License. A copy of the MIT License can be found in the [license](license) file within this repository. \n\n**Please note:** The MIT License does not grant permission to use the trade names, trademarks, service marks, or product names of the project, except as required for reasonable and customary use in describing the origin of the work and reproducing the content of the NOTICE file.\n\n### Trademarks\n\nThis project is owned and maintained by Task Venture Capital GmbH. The names and logos associated with Task Venture Capital GmbH and any related products or services are trademarks of Task Venture Capital GmbH and are not included within the scope of the MIT license granted herein. Use of these trademarks must comply with Task Venture Capital GmbH's Trademark Guidelines, and any usage must be approved in writing by Task Venture Capital GmbH.\n\n### Company Information\n\nTask Venture Capital GmbH  \nRegistered at District court Bremen HRB 35230 HB, Germany\n\nFor any legal inquiries or if you require further information, please contact us via email at hello@task.vc.\n\nBy using this repository, you acknowledge that you have read this section, agree to comply with its terms, and understand that the licensing of the code does not imply endorsement by Task Venture Capital GmbH of any derivative works.\n"
  },
  "@ship.zone/szci": {
    "npmGlobalTools": []
  }
 }
--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
 {
  "name": "@push.rocks/smartai",
-  "version": "0.7.7",
+  "version": "0.12.0",
  "private": false,
  "description": "SmartAi is a versatile TypeScript library designed to facilitate integration and interaction with various AI models, offering functionalities for chat, audio generation, document processing, and vision tasks.",
  "main": "dist_ts/index.js",
@@ -11,29 +11,29 @@
  "scripts": {
    "test": "(tstest test/ --web --verbose)",
    "typecheck": "tsbuild check",
-    "build": "(tsbuild --web --allowimplicitany)",
+    "build": "(tsbuild tsfolders --allowimplicitany)",
    "buildDocs": "(tsdoc)"
  },
  "devDependencies": {
-    "@git.zone/tsbuild": "^2.6.8",
+    "@git.zone/tsbuild": "^4.1.2",
-    "@git.zone/tsbundle": "^2.5.1",
+    "@git.zone/tsbundle": "^2.8.1",
-    "@git.zone/tsrun": "^1.3.3",
+    "@git.zone/tsrun": "^2.0.1",
-    "@git.zone/tstest": "^2.3.8",
+    "@git.zone/tstest": "^3.1.6",
    "@push.rocks/qenv": "^6.1.3",
-    "@push.rocks/tapbundle": "^6.0.3",
+    "@types/node": "^25.0.9",
    "@types/node": "^22.15.17",
    "typescript": "^5.9.3"
  },
  "dependencies": {
-    "@anthropic-ai/sdk": "^0.65.0",
+    "@anthropic-ai/sdk": "^0.71.2",
    "@mistralai/mistralai": "^1.12.0",
    "@push.rocks/smartarray": "^1.1.0",
-    "@push.rocks/smartfile": "^11.2.7",
+    "@push.rocks/smartfs": "^1.3.1",
    "@push.rocks/smartpath": "^6.0.0",
    "@push.rocks/smartpdf": "^4.1.1",
    "@push.rocks/smartpromise": "^4.2.3",
-    "@push.rocks/smartrequest": "^4.3.1",
+    "@push.rocks/smartrequest": "^5.0.1",
    "@push.rocks/webstream": "^1.0.10",
-    "openai": "^5.12.2"
+    "openai": "^6.16.0"
  },
  "repository": {
    "type": "git",
@@ -86,7 +86,8 @@
    "onlyBuiltDependencies": [
      "esbuild",
      "puppeteer"
-    ]
+    ],
    "overrides": {}
  },
  "packageManager": "pnpm@10.7.0+sha512.6b865ad4b62a1d9842b61d674a393903b871d9244954f652b8842c2b553c72176b278f64c463e52d40fff8aba385c235c8c9ecf5cc7de4fd78b8bb6d49633ab6"
 }
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
--- a/readme.hints.md
+++ b/readme.hints.md
@@ -1 +1,274 @@
 # SmartAI Project Hints
 ## Dependencies
 - Uses `@git.zone/tstest` v3.x for testing (import from `@git.zone/tstest/tapbundle`)
 - `@push.rocks/smartfs` v1.x for file system operations (replaced smartfile)
 - `@anthropic-ai/sdk` v0.71.x with extended thinking support
 - `@mistralai/mistralai` v1.x for Mistral OCR and chat capabilities
 - `openai` v6.x for OpenAI API integration
 - `@push.rocks/smartrequest` v5.x - uses `response.stream()` + `Readable.fromWeb()` for streaming
 ## Important Notes
 - When extended thinking is enabled, temperature parameter must NOT be set (or set to 1)
 - The `streamNode()` method was removed in smartrequest v5, use `response.stream()` with `Readable.fromWeb()` instead
 ## Mistral Provider Integration
 ### Overview
 The Mistral provider supports:
 - **Document AI** via Mistral OCR 3 (December 2025) - native PDF processing without image conversion
 - **Chat capabilities** using Mistral's chat models (`mistral-large-latest`, etc.)
 ### Key Advantage: Native PDF Support
 Unlike other providers that require converting PDFs to images (using SmartPdf), Mistral OCR natively accepts PDF documents as base64-encoded data. This makes document processing potentially faster and more accurate for text extraction.
 ### Configuration
 ```typescript
 import * as smartai from '@push.rocks/smartai';
 const provider = new smartai.MistralProvider({
  mistralToken: 'your-token-here',
  chatModel: 'mistral-large-latest',  // default
  ocrModel: 'mistral-ocr-latest',     // default
  tableFormat: 'markdown',             // 'markdown' or 'html'
 });
 await provider.start();
 ```
 ### Supported Methods
 | Method | Support | Notes |
 |--------|---------|-------|
 | `chat()` | ✅ | Standard chat completion |
 | `chatStream()` | ✅ | Streaming chat responses |
 | `document()` | ✅ | Native PDF OCR - no image conversion needed |
 | `vision()` | ✅ | Image OCR with optional chat analysis |
 | `audio()` | ❌ | Not supported - use ElevenLabs |
 | `research()` | ❌ | Not supported - use Perplexity |
 | `imageGenerate()` | ❌ | Not supported - use OpenAI |
 | `imageEdit()` | ❌ | Not supported - use OpenAI |
 ### Document Processing
 The `document()` method uses Mistral OCR to extract text from PDFs, then uses Mistral chat to process the user's query with the extracted content.
 ```typescript
 const result = await provider.document({
  systemMessage: 'You are a document analyst.',
  userMessage: 'Summarize this document.',
  pdfDocuments: [pdfBuffer],
  messageHistory: [],
 });
 ```
 ### API Key
 Tests require `MISTRAL_API_KEY` in `.nogit/env.json`.
 ### Pricing (as of December 2025)
 - OCR: $2 per 1,000 pages ($1 with Batch API)
 - Chat: Varies by model (see Mistral pricing page)
 ## Anthropic Extended Thinking Feature
 ### Overview
 The Anthropic provider now supports extended thinking by default across all methods. Extended thinking enables Claude to spend more time reasoning about complex problems before generating responses, leading to higher quality answers for difficult questions.
 ### Configuration
 Extended thinking is configured at the provider level during instantiation:
 ```typescript
 import * as smartai from '@push.rocks/smartai';
 const provider = new smartai.AnthropicProvider({
  anthropicToken: 'your-token-here',
  extendedThinking: 'normal', // Options: 'quick' | 'normal' | 'deep' | 'off'
 });
 ```
 ### Thinking Modes
 The `extendedThinking` parameter accepts four modes:
 | Mode       | Budget Tokens | Use Case                                        |
 | ---------- | ------------- | ----------------------------------------------- |
 | `'quick'`  | 2,048         | Lightweight reasoning for simple queries        |
 | `'normal'` | 8,000         | **Default** - Balanced reasoning for most tasks |
 | `'deep'`   | 16,000        | Complex reasoning for difficult problems        |
 | `'off'`    | 0             | Disable extended thinking                       |
 **Default Behavior**: If `extendedThinking` is not specified, it defaults to `'normal'` mode (8,000 tokens).
 ### Supported Methods
 Extended thinking is automatically applied to all Anthropic provider methods:
 - `chat()` - Synchronous chat
 - `chatStream()` - Streaming chat
 - `vision()` - Image analysis
 - `document()` - PDF document processing
 - `research()` - Web research with citations
 ### Token Budget Constraints
 **Important**: The thinking budget must be less than `max_tokens` for the API call. The current `max_tokens` values are:
 - `chatStream()`: 20,000 tokens (sufficient for all modes ✓)
 - `chat()`: 20,000 tokens (sufficient for all modes ✓)
 - `vision()`: 10,000 tokens (sufficient for all modes ✓)
 - `document()`: 20,000 tokens (sufficient for all modes ✓)
 - `research()`: 20,000 tokens for all searchDepth levels (sufficient ✓)
 ### Performance and Cost Implications
 **Token Usage**:
 - You are charged for the **full thinking tokens** generated, not just the summary
 - Higher thinking budgets may result in more thorough reasoning but increased costs
 - The budget is a **target**, not a strict limit - actual usage may vary
 **Response Quality**:
 - `'quick'`: Fast responses, basic reasoning
 - `'normal'`: Good balance between quality and speed (recommended for most use cases)
 - `'deep'`: Highest quality reasoning for complex problems, slower responses
 **Recommendations**:
 - Start with `'normal'` (default) for general usage
 - Use `'deep'` for complex analytical tasks, philosophy, mathematics, or research
 - Use `'quick'` for simple factual queries where deep reasoning isn't needed
 - Use `'off'` only if you want traditional Claude behavior without extended thinking
 ### Usage Examples
 #### Example 1: Default (Normal Mode)
 ```typescript
 const provider = new smartai.AnthropicProvider({
  anthropicToken: process.env.ANTHROPIC_TOKEN,
  // extendedThinking defaults to 'normal'
 });
 await provider.start();
 const response = await provider.chat({
  systemMessage: 'You are a helpful assistant.',
  userMessage: 'Explain the implications of quantum computing.',
  messageHistory: [],
 });
 ```
 #### Example 2: Deep Thinking for Complex Analysis
 ```typescript
 const provider = new smartai.AnthropicProvider({
  anthropicToken: process.env.ANTHROPIC_TOKEN,
  extendedThinking: 'deep', // 16,000 token budget
 });
 await provider.start();
 const response = await provider.chat({
  systemMessage: 'You are a philosopher and ethicist.',
  userMessage: 'Analyze the trolley problem from multiple ethical frameworks.',
  messageHistory: [],
 });
 ```
 #### Example 3: Quick Mode for Simple Queries
 ```typescript
 const provider = new smartai.AnthropicProvider({
  anthropicToken: process.env.ANTHROPIC_TOKEN,
  extendedThinking: 'quick', // 2,048 token budget
 });
 await provider.start();
 const response = await provider.chat({
  systemMessage: 'You are a helpful assistant.',
  userMessage: 'What is the capital of France?',
  messageHistory: [],
 });
 ```
 #### Example 4: Disable Thinking
 ```typescript
 const provider = new smartai.AnthropicProvider({
  anthropicToken: process.env.ANTHROPIC_TOKEN,
  extendedThinking: 'off', // No extended thinking
 });
 await provider.start();
 const response = await provider.chat({
  systemMessage: 'You are a helpful assistant.',
  userMessage: 'Tell me a joke.',
  messageHistory: [],
 });
 ```
 #### Example 5: Extended Thinking with Vision
 ```typescript
 const provider = new smartai.AnthropicProvider({
  anthropicToken: process.env.ANTHROPIC_TOKEN,
  extendedThinking: 'normal',
 });
 await provider.start();
 const imageBuffer = await fs.promises.readFile('./image.jpg');
 const analysis = await provider.vision({
  image: imageBuffer,
  prompt: 'Analyze this image in detail and explain what you see.',
 });
 ```
 ### Testing
 Comprehensive tests for extended thinking are available in:
 - `test/test.thinking.anthropic.ts` - Tests all thinking modes
 Run tests with:
 ```bash
 pnpm test
 ```
 Run specific thinking tests:
 ```bash
 npx tstest test/test.thinking.anthropic.ts --verbose
 ```
 ### API Reference
 According to Anthropic's documentation:
 - Extended thinking is supported on Claude Sonnet 4.5, 4, 3.7, Haiku 4.5, and Opus 4.1, 4
 - The current model used is `claude-sonnet-4-5-20250929`
 - Minimum thinking budget is 1,024 tokens
 - Thinking budget must be less than `max_tokens`
 ### Implementation Details
 The extended thinking feature is implemented via:
 1. **Interface**: `IAnthropicProviderOptions.extendedThinking` property
 2. **Helper Method**: `getThinkingConfig()` private method that maps modes to token budgets
 3. **API Parameter**: Adds `thinking: { type: 'enabled', budget_tokens: number }` to all API calls
 The thinking configuration is applied automatically to all API calls when the provider is instantiated.
--- a/readme.md
+++ b/readme.md
@@ -1,11 +1,16 @@
 # @push.rocks/smartai
 **One API to rule them all** 🚀
 [![npm version](https://img.shields.io/npm/v/@push.rocks/smartai.svg)](https://www.npmjs.com/package/@push.rocks/smartai)
 [![TypeScript](https://img.shields.io/badge/TypeScript-5.x-blue.svg)](https://www.typescriptlang.org/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-SmartAI unifies the world's leading AI providers - OpenAI, Anthropic, Perplexity, Ollama, Groq, XAI, Exo, and ElevenLabs - under a single, elegant TypeScript interface. Build AI applications at lightning speed without vendor lock-in. 
+SmartAI unifies the world's leading AI providers - OpenAI, Anthropic, Mistral, Perplexity, Ollama, Groq, XAI, Exo, and ElevenLabs - under a single, elegant TypeScript interface. Build AI applications at lightning speed without vendor lock-in.
 ## Issue Reporting and Security
 For reporting bugs, issues, or security vulnerabilities, please visit [community.foss.global/](https://community.foss.global/). This is the central community hub for all issue reporting. Developers who sign and comply with our contribution agreement and go through identification can also get a [code.foss.global/](https://code.foss.global/) account to submit Pull Requests directly.
 ## 🎯 Why SmartAI?
@@ -31,8 +36,8 @@ const ai = new SmartAi({
  anthropicToken: 'sk-ant-...',
  elevenlabsToken: 'sk-...',
  elevenlabs: {
-    defaultVoiceId: '19STyYD15bswVz51nqLf' // Optional: Samara voice
+    defaultVoiceId: '19STyYD15bswVz51nqLf', // Optional: Samara voice
-  }
+  },
 });
 await ai.start();
@@ -41,7 +46,7 @@ await ai.start();
 const response = await ai.openaiProvider.chat({
  systemMessage: 'You are a helpful assistant.',
  userMessage: 'Explain quantum computing in simple terms',
-  messageHistory: []
+  messageHistory: [],
 });
 ```
@@ -50,9 +55,10 @@ const response = await ai.openaiProvider.chat({
 Choose the right provider for your use case:
 | Provider       | Chat | Streaming | TTS | Vision | Documents | Research | Images | Highlights                                                      |
-|----------|:----:|:---------:|:---:|:------:|:---------:|:--------:|:------:|------------|
+| -------------- | :--: | :-------: | :-: | :----: | :-------: | :------: | :----: | --------------------------------------------------------------- |
 | **OpenAI**     |  ✅  |    ✅     | ✅  |   ✅   |    ✅     |    ✅    |   ✅   | • gpt-image-1<br>• DALL-E 3<br>• Deep research API              |
 | **Anthropic**  |  ✅  |    ✅     | ❌  |   ✅   |    ✅     |    ✅    |   ❌   | • Claude Sonnet 4.5<br>• Superior reasoning<br>• Web search API |
 | **Mistral**    |  ✅  |    ✅     | ❌  |   ✅   |    ✅     |    ❌    |   ❌   | • Native PDF OCR<br>• mistral-large<br>• Fast inference         |
 | **ElevenLabs** |  ❌  |    ❌     | ✅  |   ❌   |    ❌     |    ❌    |   ❌   | • Premium TTS<br>• 70+ languages<br>• Natural voices            |
 | **Ollama**     |  ✅  |    ✅     | ❌  |   ✅   |    ✅     |    ❌    |   ❌   | • 100% local<br>• Privacy-first<br>• No API costs               |
 | **XAI**        |  ✅  |    ✅     | ❌  |   ❌   |    ✅     |    ❌    |   ❌   | • Grok models<br>• Real-time data<br>• Uncensored               |
@@ -71,21 +77,21 @@ Works identically across all providers:
 const gptResponse = await ai.openaiProvider.chat({
  systemMessage: 'You are a expert physicist.',
  userMessage: 'Explain the implications of quantum entanglement',
-  messageHistory: []
+  messageHistory: [],
 });
 // Use Claude for safety-critical applications
 const claudeResponse = await ai.anthropicProvider.chat({
  systemMessage: 'You are a medical advisor.',
  userMessage: 'Review this patient data for concerns',
-  messageHistory: []
+  messageHistory: [],
 });
 // Use Groq for lightning-fast responses
 const groqResponse = await ai.groqProvider.chat({
  systemMessage: 'You are a code reviewer.',
  userMessage: 'Quick! Find the bug in this code: ...',
-  messageHistory: []
+  messageHistory: [],
 });
 ```
@@ -115,7 +121,7 @@ Generate natural voices with OpenAI or ElevenLabs:
 ```typescript
 // OpenAI TTS
 const audioStream = await ai.openaiProvider.audio({
-  message: 'Welcome to the future of AI development!'
+  message: 'Welcome to the future of AI development!',
 });
 // ElevenLabs TTS - Premium quality, natural voices (uses v3 by default)
@@ -123,12 +129,13 @@ const elevenLabsAudio = await ai.elevenlabsProvider.audio({
  message: 'Experience the most lifelike text to speech technology.',
  voiceId: '19STyYD15bswVz51nqLf', // Optional: Samara voice
  modelId: 'eleven_v3', // Optional: defaults to eleven_v3 (70+ languages, most expressive)
-  voiceSettings: { // Optional: fine-tune voice characteristics
+  voiceSettings: {
    // Optional: fine-tune voice characteristics
    stability: 0.5, // 0-1: Speech consistency
    similarity_boost: 0.8, // 0-1: Voice similarity to original
    style: 0.0, // 0-1: Expressiveness (higher = more expressive)
-    use_speaker_boost: true // Enhanced clarity
+    use_speaker_boost: true, // Enhanced clarity
-  }
+  },
 });
 // Stream directly to speakers
@@ -148,19 +155,19 @@ const image = fs.readFileSync('product-photo.jpg');
 // OpenAI: General purpose vision
 const gptVision = await ai.openaiProvider.vision({
  image,
-  prompt: 'Describe this product and suggest marketing angles'
+  prompt: 'Describe this product and suggest marketing angles',
 });
 // Anthropic: Detailed analysis
 const claudeVision = await ai.anthropicProvider.vision({
  image,
-  prompt: 'Identify any safety concerns or defects'
+  prompt: 'Identify any safety concerns or defects',
 });
 // Ollama: Private, local analysis
 const ollamaVision = await ai.ollamaProvider.vision({
  image,
-  prompt: 'Extract all text and categorize the content'
+  prompt: 'Extract all text and categorize the content',
 });
 ```
@@ -177,7 +184,7 @@ const analysis = await ai.openaiProvider.document({
  systemMessage: 'You are a legal expert.',
  userMessage: 'Compare these documents and highlight key differences',
  messageHistory: [],
-  pdfDocuments: [contract, invoice]
+  pdfDocuments: [contract, invoice],
 });
 // Multi-document analysis
@@ -186,7 +193,7 @@ const taxAnalysis = await ai.anthropicProvider.document({
  systemMessage: 'You are a tax advisor.',
  userMessage: 'Prepare a tax summary from these documents',
  messageHistory: [],
-  pdfDocuments: taxDocs
+  pdfDocuments: taxDocs,
 });
 ```
@@ -199,7 +206,7 @@ Perform deep research with web search capabilities across multiple providers:
 const deepResearch = await ai.openaiProvider.research({
  query: 'What are the latest developments in quantum computing?',
  searchDepth: 'deep',
-  includeWebSearch: true
+  includeWebSearch: true,
 });
 console.log(deepResearch.answer);
@@ -209,32 +216,105 @@ console.log('Sources:', deepResearch.sources);
 const anthropic = new AnthropicProvider({
  anthropicToken: 'sk-ant-...',
  enableWebSearch: true,
-  searchDomainAllowList: ['nature.com', 'science.org']
+  searchDomainAllowList: ['nature.com', 'science.org'],
 });
 const scientificResearch = await anthropic.research({
  query: 'Latest breakthroughs in CRISPR gene editing',
-  searchDepth: 'advanced'
+  searchDepth: 'advanced',
 });
 // Perplexity - Research-focused with citations
 const perplexityResearch = await ai.perplexityProvider.research({
  query: 'Current state of autonomous vehicle technology',
-  searchDepth: 'deep' // Uses Sonar Pro model
+  searchDepth: 'deep', // Uses Sonar Pro model
 });
 ```
 **Research Options:**
 - `searchDepth`: 'basic' | 'advanced' | 'deep'
 - `maxSources`: Number of sources to include
 - `includeWebSearch`: Enable web search (OpenAI)
 - `background`: Run as background task (OpenAI)
 **Supported Providers:**
 - **OpenAI**: Deep Research API with specialized models (`o3-deep-research-2025-06-26`, `o4-mini-deep-research-2025-06-26`)
 - **Anthropic**: Web Search API with domain filtering
 - **Perplexity**: Sonar and Sonar Pro models with built-in citations
 ### 🧠 Extended Thinking (Anthropic)
 Enable Claude to spend more time reasoning about complex problems before generating responses:
 ```typescript
 import { AnthropicProvider } from '@push.rocks/smartai';
 // Configure extended thinking mode at provider level
 const anthropic = new AnthropicProvider({
  anthropicToken: 'sk-ant-...',
  extendedThinking: 'normal', // Options: 'quick' | 'normal' | 'deep' | 'off'
 });
 await anthropic.start();
 // Extended thinking is automatically applied to all methods
 const response = await anthropic.chat({
  systemMessage: 'You are an expert mathematician.',
  userMessage: 'Prove the Pythagorean theorem from first principles',
  messageHistory: [],
 });
 ```
 **Thinking Modes:**
 | Mode | Budget Tokens | Use Case |
 |------|---------------|----------|
 | `'quick'` | 2,048 | Lightweight reasoning for simple queries |
 | `'normal'` | 8,000 | **Default** - Balanced reasoning for most tasks |
 | `'deep'` | 16,000 | Complex reasoning for difficult problems |
 | `'off'` | 0 | Disable extended thinking |
 **Best Practices:**
 - Start with `'normal'` (default) for general usage
 - Use `'deep'` for complex analytical tasks, philosophy, mathematics, or research
 - Use `'quick'` for simple factual queries where deep reasoning isn't needed
 - Thinking budget counts against total token usage
 ### 📑 Native PDF OCR (Mistral)
 Mistral provides native PDF document processing via their OCR API - no image conversion required:
 ```typescript
 import { MistralProvider } from '@push.rocks/smartai';
 const mistral = new MistralProvider({
  mistralToken: 'your-api-key',
  chatModel: 'mistral-large-latest',  // Default
  ocrModel: 'mistral-ocr-latest',     // Default
  tableFormat: 'markdown',             // 'markdown' | 'html'
 });
 await mistral.start();
 // Direct PDF processing - no image conversion overhead
 const result = await mistral.document({
  systemMessage: 'You are a document analyst.',
  userMessage: 'Extract all invoice details and calculate the total.',
  pdfDocuments: [invoicePdfBuffer],
  messageHistory: [],
 });
 ```
 **Key Advantage**: Unlike other providers that convert PDFs to images first, Mistral's OCR API processes PDFs natively, potentially offering faster and more accurate text extraction for document-heavy workloads.
 **Supported Formats:**
 - Native PDF processing via Files API
 - Image OCR (JPEG, PNG, GIF, WebP) for vision tasks
 - Table extraction with markdown or HTML output
 ### 🎨 Image Generation & Editing
 Generate and edit images with OpenAI's cutting-edge models:
@@ -245,7 +325,7 @@ const image = await ai.openaiProvider.imageGenerate({
  prompt: 'A futuristic robot assistant in a modern office, digital art',
  model: 'gpt-image-1',
  quality: 'high',
-  size: '1024x1024'
+  size: '1024x1024',
 });
 // Save the generated image
@@ -259,7 +339,7 @@ const logo = await ai.openaiProvider.imageGenerate({
  quality: 'high',
  size: '1024x1024',
  background: 'transparent',
-  outputFormat: 'png'
+  outputFormat: 'png',
 });
 // WebP with compression for web use
@@ -269,15 +349,16 @@ const webImage = await ai.openaiProvider.imageGenerate({
  quality: 'high',
  size: '1536x1024',
  outputFormat: 'webp',
-  outputCompression: 85
+  outputCompression: 85,
 });
 // Superior text rendering (gpt-image-1's strength)
 const signage = await ai.openaiProvider.imageGenerate({
-  prompt: 'Vintage cafe sign saying "COFFEE & CODE" in hand-lettered typography',
+  prompt:
    'Vintage cafe sign saying "COFFEE & CODE" in hand-lettered typography',
  model: 'gpt-image-1',
  quality: 'high',
-  size: '1024x1024'
+  size: '1024x1024',
 });
 // Generate multiple variations at once
@@ -286,7 +367,7 @@ const variations = await ai.openaiProvider.imageGenerate({
  model: 'gpt-image-1',
  n: 3,
  quality: 'medium',
-  size: '1024x1024'
+  size: '1024x1024',
 });
 // Edit an existing image
@@ -294,11 +375,12 @@ const editedImage = await ai.openaiProvider.imageEdit({
  image: originalImageBuffer,
  prompt: 'Add sunglasses and change the background to a beach sunset',
  model: 'gpt-image-1',
-  quality: 'high'
+  quality: 'high',
 });
 ```
 **Image Generation Options:**
 - `model`: 'gpt-image-1' | 'dall-e-3' | 'dall-e-2'
 - `quality`: 'low' | 'medium' | 'high' | 'auto'
 - `size`: Multiple aspect ratios up to 4096×4096
@@ -309,6 +391,7 @@ const editedImage = await ai.openaiProvider.imageEdit({
 - `n`: Number of images (1-10)
 **gpt-image-1 Advantages:**
 - Superior text rendering in images
 - Up to 4096×4096 resolution
 - Transparent background support
@@ -365,7 +448,7 @@ async function handleCustomerQuery(query: string, history: ChatMessage[]) {
 ```typescript
 const codeReviewer = new SmartAi({
-  groqToken: process.env.GROQ_KEY // Groq for speed
+  groqToken: process.env.GROQ_KEY, // Groq for speed
 });
 async function reviewCode(code: string, language: string) {
@@ -378,7 +461,7 @@ async function reviewCode(code: string, language: string) {
                    - Best practices
                    - Potential bugs`,
    userMessage: `Review this code:\n\n${code}`,
-    messageHistory: []
+    messageHistory: [],
  });
  console.log(`Review completed in ${Date.now() - startTime}ms`);
@@ -390,15 +473,16 @@ async function reviewCode(code: string, language: string) {
 ```typescript
 const researcher = new SmartAi({
-  perplexityToken: process.env.PERPLEXITY_KEY
+  perplexityToken: process.env.PERPLEXITY_KEY,
 });
 async function research(topic: string) {
  // Perplexity excels at web-aware research
  const findings = await researcher.perplexityProvider.chat({
-    systemMessage: 'You are a research assistant. Provide factual, cited information.',
+    systemMessage:
      'You are a research assistant. Provide factual, cited information.',
    userMessage: `Research the latest developments in ${topic}`,
-    messageHistory: []
+    messageHistory: [],
  });
  return findings.message;
@@ -412,8 +496,8 @@ const localAI = new SmartAi({
  ollama: {
    baseUrl: 'http://localhost:11434',
    model: 'llama2',
-    visionModel: 'llava'
+    visionModel: 'llava',
-  }
+  },
 });
 // Process sensitive documents without leaving your infrastructure
@@ -422,7 +506,7 @@ async function analyzeSensitiveDoc(pdfBuffer: Buffer) {
    systemMessage: 'Extract and summarize key information.',
    userMessage: 'Analyze this confidential document',
    messageHistory: [],
-    pdfDocuments: [pdfBuffer]
+    pdfDocuments: [pdfBuffer],
  });
  // Data never leaves your servers
@@ -464,7 +548,9 @@ class SmartAIRouter {
 ```typescript
 // Don't wait for the entire response
 async function streamResponse(userQuery: string) {
-  const stream = await ai.openaiProvider.chatStream(createInputStream(userQuery));
+  const stream = await ai.openaiProvider.chatStream(
    createInputStream(userQuery),
  );
  // Process tokens as they arrive
  for await (const chunk of stream) {
@@ -500,7 +586,7 @@ const translationStream = new TransformStream({
  async transform(chunk, controller) {
    const translated = await translateChunk(chunk);
    controller.enqueue(translated);
-  }
+  },
 });
 const responseStream = await ai.openaiProvider.chatStream(input);
@@ -540,7 +626,7 @@ class UsageTracker {
      provider,
      duration: Date.now() - start,
      inputTokens: estimateTokens(options),
-      outputTokens: estimateTokens(response.message)
+      outputTokens: estimateTokens(response.message),
    };
    await this.logUsage(usage);
@@ -589,9 +675,10 @@ export ELEVENLABS_API_KEY=sk-...
 ## 🎯 Choosing the Right Provider
 | Use Case              | Recommended Provider | Why                                                       |
-|----------|---------------------|-----|
+| --------------------- | -------------------- | --------------------------------------------------------- |
 | **General Purpose**   | OpenAI               | Most features, stable, well-documented                    |
 | **Complex Reasoning** | Anthropic            | Superior logical thinking, safer outputs                  |
 | **Document OCR**      | Mistral              | Native PDF processing, no image conversion overhead       |
 | **Research & Facts**  | Perplexity           | Web-aware, provides citations                             |
 | **Deep Research**     | OpenAI               | Deep Research API with comprehensive analysis             |
 | **Premium TTS**       | ElevenLabs           | Most natural voices, 70+ languages, superior quality (v3) |
@@ -613,19 +700,21 @@ export ELEVENLABS_API_KEY=sk-...
 ## License and Legal Information
-This repository contains open-source code that is licensed under the MIT License. A copy of the MIT License can be found in the [license](license) file within this repository. 
+This repository contains open-source code licensed under the MIT License. A copy of the license can be found in the [LICENSE](./LICENSE) file.
 **Please note:** The MIT License does not grant permission to use the trade names, trademarks, service marks, or product names of the project, except as required for reasonable and customary use in describing the origin of the work and reproducing the content of the NOTICE file.
 ### Trademarks
-This project is owned and maintained by Task Venture Capital GmbH. The names and logos associated with Task Venture Capital GmbH and any related products or services are trademarks of Task Venture Capital GmbH and are not included within the scope of the MIT license granted herein. Use of these trademarks must comply with Task Venture Capital GmbH's Trademark Guidelines, and any usage must be approved in writing by Task Venture Capital GmbH.
+This project is owned and maintained by Task Venture Capital GmbH. The names and logos associated with Task Venture Capital GmbH and any related products or services are trademarks of Task Venture Capital GmbH or third parties, and are not included within the scope of the MIT license granted herein.
 Use of these trademarks must comply with Task Venture Capital GmbH's Trademark Guidelines or the guidelines of the respective third-party owners, and any usage must be approved in writing. Third-party trademarks used herein are the property of their respective owners and used only in a descriptive manner, e.g. for an implementation of an API or similar.
 ### Company Information
 Task Venture Capital GmbH
-Registered at District court Bremen HRB 35230 HB, Germany
+Registered at District Court Bremen HRB 35230 HB, Germany
-For any legal inquiries or if you require further information, please contact us via email at hello@task.vc.
+For any legal inquiries or further information, please contact us via email at hello@task.vc.
 By using this repository, you acknowledge that you have read this section, agree to comply with its terms, and understand that the licensing of the code does not imply endorsement by Task Venture Capital GmbH of any derivative works.
--- a/test/test.audio.elevenlabs.ts
+++ b/test/test.audio.elevenlabs.ts
@@ -1,8 +1,9 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
-import * as smartfile from '@push.rocks/smartfile';
+import { SmartFs, SmartFsProviderNode } from '@push.rocks/smartfs';
 const testQenv = new qenv.Qenv('./', './.nogit/');
 const smartfs = new SmartFs(new SmartFsProviderNode());
 import * as smartai from '../ts/index.js';
@@ -27,7 +28,7 @@ tap.test('ElevenLabs Audio: should create audio response', async () => {
    chunks.push(chunk as Uint8Array);
  }
  const audioBuffer = Buffer.concat(chunks);
-  await smartfile.fs.toFs(audioBuffer, './.nogit/testoutput_elevenlabs.mp3');
+  await smartfs.file('./.nogit/testoutput_elevenlabs.mp3').write(audioBuffer);
  console.log(`Audio Buffer length: ${audioBuffer.length}`);
  expect(audioBuffer.length).toBeGreaterThan(0);
 });
@@ -42,7 +43,7 @@ tap.test('ElevenLabs Audio: should create audio with custom voice', async () =>
    chunks.push(chunk as Uint8Array);
  }
  const audioBuffer = Buffer.concat(chunks);
-  await smartfile.fs.toFs(audioBuffer, './.nogit/testoutput_elevenlabs_custom.mp3');
+  await smartfs.file('./.nogit/testoutput_elevenlabs_custom.mp3').write(audioBuffer);
  console.log(`Audio Buffer length (custom voice): ${audioBuffer.length}`);
  expect(audioBuffer.length).toBeGreaterThan(0);
 });
--- a/test/test.audio.openai.ts
+++ b/test/test.audio.openai.ts
@@ -1,8 +1,9 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
-import * as smartfile from '@push.rocks/smartfile';
+import { SmartFs, SmartFsProviderNode } from '@push.rocks/smartfs';
 const testQenv = new qenv.Qenv('./', './.nogit/');
 const smartfs = new SmartFs(new SmartFsProviderNode());
 import * as smartai from '../ts/index.js';
@@ -26,7 +27,7 @@ tap.test('OpenAI Audio: should create audio response', async () => {
    chunks.push(chunk as Uint8Array);
  }
  const audioBuffer = Buffer.concat(chunks);
-  await smartfile.fs.toFs(audioBuffer, './.nogit/testoutput.mp3');
+  await smartfs.file('./.nogit/testoutput.mp3').write(audioBuffer);
  console.log(`Audio Buffer length: ${audioBuffer.length}`);
  // Assert that the resulting buffer is not empty.
  expect(audioBuffer.length).toBeGreaterThan(0);
--- a/test/test.audio.stubs.ts
+++ b/test/test.audio.stubs.ts
@@ -1,4 +1,4 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 const testQenv = new qenv.Qenv('./', './.nogit/');
--- a/test/test.basic.ts
+++ b/test/test.basic.ts
@@ -1,4 +1,4 @@
-import { tap, expect } from '@push.rocks/tapbundle';
+import { tap, expect } from '@git.zone/tstest/tapbundle';
 import * as smartai from '../ts/index.js';
 // Basic instantiation tests that don't require API tokens
--- a/test/test.chat.anthropic.ts
+++ b/test/test.chat.anthropic.ts
@@ -1,4 +1,4 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 const testQenv = new qenv.Qenv('./', './.nogit/');
--- a/test/test.chat.mistral.ts
+++ b/test/test.chat.mistral.ts
@@ -0,0 +1,66 @@
 import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 const testQenv = new qenv.Qenv('./', './.nogit/');
 import * as smartai from '../ts/index.js';
 let mistralProvider: smartai.MistralProvider;
 tap.test('Mistral Chat: should create and start Mistral provider', async () => {
  mistralProvider = new smartai.MistralProvider({
    mistralToken: await testQenv.getEnvVarOnDemand('MISTRAL_API_KEY'),
  });
  await mistralProvider.start();
  expect(mistralProvider).toBeInstanceOf(smartai.MistralProvider);
 });
 tap.test('Mistral Chat: should create chat response', async () => {
  const userMessage = 'What is the capital of France? Answer in one word.';
  const response = await mistralProvider.chat({
    systemMessage: 'You are a helpful assistant. Be concise.',
    userMessage: userMessage,
    messageHistory: [],
  });
  console.log(`Mistral Chat - User: ${userMessage}`);
  console.log(`Mistral Chat - Response: ${response.message}`);
  expect(response.role).toEqual('assistant');
  expect(response.message).toBeTruthy();
  expect(response.message.toLowerCase()).toInclude('paris');
 });
 tap.test('Mistral Chat: should handle message history', async () => {
  const messageHistory: smartai.ChatMessage[] = [
    { role: 'user', content: 'My name is Claude Test' },
    { role: 'assistant', content: 'Nice to meet you, Claude Test!' }
  ];
  const response = await mistralProvider.chat({
    systemMessage: 'You are a helpful assistant with good memory.',
    userMessage: 'What is my name?',
    messageHistory: messageHistory,
  });
  console.log(`Mistral Memory Test - Response: ${response.message}`);
  expect(response.message.toLowerCase()).toInclude('claude test');
 });
 tap.test('Mistral Chat: should handle longer conversations', async () => {
  const response = await mistralProvider.chat({
    systemMessage: 'You are a helpful coding assistant.',
    userMessage: 'Write a simple hello world function in TypeScript. Keep it brief.',
    messageHistory: [],
  });
  console.log(`Mistral Coding Test - Response: ${response.message}`);
  expect(response.message).toBeTruthy();
  // Should contain some TypeScript/function code
  expect(response.message).toInclude('function');
 });
 tap.test('Mistral Chat: should stop the provider', async () => {
  await mistralProvider.stop();
 });
 export default tap.start();
--- a/test/test.chat.openai.ts
+++ b/test/test.chat.openai.ts
@@ -1,4 +1,4 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 const testQenv = new qenv.Qenv('./', './.nogit/');
--- a/test/test.document.anthropic.ts
+++ b/test/test.document.anthropic.ts
@@ -1,9 +1,10 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 import * as smartrequest from '@push.rocks/smartrequest';
-import * as smartfile from '@push.rocks/smartfile';
+import { SmartFs, SmartFsProviderNode } from '@push.rocks/smartfs';
 const testQenv = new qenv.Qenv('./', './.nogit/');
 const smartfs = new SmartFs(new SmartFsProviderNode());
 import * as smartai from '../ts/index.js';
@@ -41,7 +42,7 @@ tap.test('Anthropic Document: should handle complex document analysis', async ()
  let pdfBuffer: Uint8Array;
  try {
-    pdfBuffer = await smartfile.fs.toBuffer(pdfPath);
+    pdfBuffer = await smartfs.file(pdfPath).read();
  } catch (error) {
    // If the file doesn't exist, use the dummy PDF
    console.log('Demo PDF not found, using dummy PDF instead');
--- a/test/test.document.mistral.ts
+++ b/test/test.document.mistral.ts
@@ -0,0 +1,100 @@
 import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 import * as smartrequest from '@push.rocks/smartrequest';
 import { SmartFs, SmartFsProviderNode } from '@push.rocks/smartfs';
 const testQenv = new qenv.Qenv('./', './.nogit/');
 const smartfs = new SmartFs(new SmartFsProviderNode());
 import * as smartai from '../ts/index.js';
 let mistralProvider: smartai.MistralProvider;
 tap.test('Mistral Document: should create and start Mistral provider', async () => {
  mistralProvider = new smartai.MistralProvider({
    mistralToken: await testQenv.getEnvVarOnDemand('MISTRAL_API_KEY'),
    tableFormat: 'markdown',
  });
  await mistralProvider.start();
  expect(mistralProvider).toBeInstanceOf(smartai.MistralProvider);
 });
 tap.test('Mistral Document: should process a PDF document', async () => {
  const pdfUrl = 'https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf';
  const pdfResponse = await smartrequest.SmartRequest.create()
    .url(pdfUrl)
    .get();
  const result = await mistralProvider.document({
    systemMessage: 'Classify the document. Only the following answers are allowed: "invoice", "bank account statement", "contract", "test document", "other". The answer should only contain the keyword for machine use.',
    userMessage: 'Classify this document.',
    messageHistory: [],
    pdfDocuments: [Buffer.from(await pdfResponse.arrayBuffer())],
  });
  console.log(`Mistral Document - Result:`, result);
  expect(result).toBeTruthy();
  expect(result.message).toBeTruthy();
 });
 tap.test('Mistral Document: should handle complex document analysis', async () => {
  // Test with the demo PDF if it exists
  const pdfPath = './.nogit/demo_without_textlayer.pdf';
  let pdfBuffer: Uint8Array;
  try {
    pdfBuffer = await smartfs.file(pdfPath).read();
  } catch (error) {
    // If the file doesn't exist, use the dummy PDF
    console.log('Demo PDF not found, using dummy PDF instead');
    const pdfUrl = 'https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf';
    const pdfResponse = await smartrequest.SmartRequest.create()
      .url(pdfUrl)
      .get();
    pdfBuffer = Buffer.from(await pdfResponse.arrayBuffer());
  }
  const result = await mistralProvider.document({
    systemMessage: `
      Analyze this document and provide a JSON response with the following structure:
      {
        "documentType": "string",
        "hasText": boolean,
        "summary": "string"
      }
    `,
    userMessage: 'Analyze this document.',
    messageHistory: [],
    pdfDocuments: [pdfBuffer],
  });
  console.log(`Mistral Complex Document Analysis:`, result);
  expect(result).toBeTruthy();
  expect(result.message).toBeTruthy();
 });
 tap.test('Mistral Document: should process multiple PDF documents', async () => {
  const pdfUrl = 'https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf';
  const pdfResponse = await smartrequest.SmartRequest.create()
    .url(pdfUrl)
    .get();
  const pdfBuffer = Buffer.from(await pdfResponse.arrayBuffer());
  const result = await mistralProvider.document({
    systemMessage: 'You are a document comparison assistant.',
    userMessage: 'Are these two documents the same? Answer yes or no.',
    messageHistory: [],
    pdfDocuments: [pdfBuffer, pdfBuffer], // Same document twice for test
  });
  console.log(`Mistral Multi-Document - Result:`, result);
  expect(result).toBeTruthy();
  expect(result.message).toBeTruthy();
 });
 tap.test('Mistral Document: should stop the provider', async () => {
  await mistralProvider.stop();
 });
 export default tap.start();
--- a/test/test.document.openai.ts
+++ b/test/test.document.openai.ts
@@ -1,9 +1,10 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 import * as smartrequest from '@push.rocks/smartrequest';
-import * as smartfile from '@push.rocks/smartfile';
+import { SmartFs, SmartFsProviderNode } from '@push.rocks/smartfs';
 const testQenv = new qenv.Qenv('./', './.nogit/');
 const smartfs = new SmartFs(new SmartFsProviderNode());
 import * as smartai from '../ts/index.js';
@@ -32,7 +33,7 @@ tap.test('OpenAI Document: should document a pdf', async () => {
 });
 tap.test('OpenAI Document: should recognize companies in a pdf', async () => {
-  const pdfBuffer = await smartfile.fs.toBuffer('./.nogit/demo_without_textlayer.pdf');
+  const pdfBuffer = await smartfs.file('./.nogit/demo_without_textlayer.pdf').read();
  const result = await testSmartai.openaiProvider.document({
    systemMessage: `
      summarize the document.
--- a/test/test.image.openai.ts
+++ b/test/test.image.openai.ts
@@ -1,4 +1,4 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 import * as smartai from '../ts/index.js';
 import * as path from 'path';
--- a/test/test.interfaces.ts
+++ b/test/test.interfaces.ts
@@ -1,4 +1,4 @@
-import { tap, expect } from '@push.rocks/tapbundle';
+import { tap, expect } from '@git.zone/tstest/tapbundle';
 import * as smartai from '../ts/index.js';
 // Test interface exports and type checking
--- a/test/test.research.anthropic.ts
+++ b/test/test.research.anthropic.ts
@@ -1,4 +1,4 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 import * as smartai from '../ts/index.js';
 import * as path from 'path';
--- a/test/test.research.openai.ts
+++ b/test/test.research.openai.ts
@@ -1,4 +1,4 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 import * as smartai from '../ts/index.js';
 import * as path from 'path';
--- a/test/test.research.stubs.ts
+++ b/test/test.research.stubs.ts
@@ -1,4 +1,4 @@
-import { tap, expect } from '@push.rocks/tapbundle';
+import { tap, expect } from '@git.zone/tstest/tapbundle';
 import * as smartai from '../ts/index.js';
 // Test research method stubs for providers without full implementation
--- a/test/test.thinking.anthropic.ts
+++ b/test/test.thinking.anthropic.ts
@@ -0,0 +1,151 @@
 import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
 const testQenv = new qenv.Qenv('./', './.nogit/');
 import * as smartai from '../ts/index.js';
 let anthropicProviderQuick: smartai.AnthropicProvider;
 let anthropicProviderNormal: smartai.AnthropicProvider;
 let anthropicProviderDeep: smartai.AnthropicProvider;
 let anthropicProviderOff: smartai.AnthropicProvider;
 // Test 'quick' mode
 tap.test('Extended Thinking: should create Anthropic provider with quick mode', async () => {
  anthropicProviderQuick = new smartai.AnthropicProvider({
    anthropicToken: await testQenv.getEnvVarOnDemand('ANTHROPIC_TOKEN'),
    extendedThinking: 'quick',
  });
  await anthropicProviderQuick.start();
  expect(anthropicProviderQuick).toBeInstanceOf(smartai.AnthropicProvider);
 });
 tap.test('Extended Thinking: should chat with quick mode (2048 tokens)', async () => {
  const userMessage = 'Explain quantum entanglement in simple terms.';
  const response = await anthropicProviderQuick.chat({
    systemMessage: 'You are a helpful physics teacher.',
    userMessage: userMessage,
    messageHistory: [],
  });
  console.log(`Quick Mode - User: ${userMessage}`);
  console.log(`Quick Mode - Response length: ${response.message.length} chars`);
  expect(response.role).toEqual('assistant');
  expect(response.message).toBeTruthy();
  expect(response.message.toLowerCase()).toInclude('quantum');
 });
 tap.test('Extended Thinking: should stop quick mode provider', async () => {
  await anthropicProviderQuick.stop();
 });
 // Test 'normal' mode (default)
 tap.test('Extended Thinking: should create Anthropic provider with normal mode (default)', async () => {
  anthropicProviderNormal = new smartai.AnthropicProvider({
    anthropicToken: await testQenv.getEnvVarOnDemand('ANTHROPIC_TOKEN'),
    // extendedThinking not specified, should default to 'normal'
  });
  await anthropicProviderNormal.start();
  expect(anthropicProviderNormal).toBeInstanceOf(smartai.AnthropicProvider);
 });
 tap.test('Extended Thinking: should chat with normal mode (8000 tokens default)', async () => {
  const userMessage = 'What are the implications of the P vs NP problem?';
  const response = await anthropicProviderNormal.chat({
    systemMessage: 'You are a helpful computer science expert.',
    userMessage: userMessage,
    messageHistory: [],
  });
  console.log(`Normal Mode - User: ${userMessage}`);
  console.log(`Normal Mode - Response length: ${response.message.length} chars`);
  expect(response.role).toEqual('assistant');
  expect(response.message).toBeTruthy();
  expect(response.message.length).toBeGreaterThan(50);
 });
 tap.test('Extended Thinking: should stop normal mode provider', async () => {
  await anthropicProviderNormal.stop();
 });
 // Test 'deep' mode
 tap.test('Extended Thinking: should create Anthropic provider with deep mode', async () => {
  anthropicProviderDeep = new smartai.AnthropicProvider({
    anthropicToken: await testQenv.getEnvVarOnDemand('ANTHROPIC_TOKEN'),
    extendedThinking: 'deep',
  });
  await anthropicProviderDeep.start();
  expect(anthropicProviderDeep).toBeInstanceOf(smartai.AnthropicProvider);
 });
 tap.test('Extended Thinking: should chat with deep mode (16000 tokens)', async () => {
  const userMessage = 'Analyze the philosophical implications of artificial consciousness.';
  const response = await anthropicProviderDeep.chat({
    systemMessage: 'You are a philosopher and cognitive scientist.',
    userMessage: userMessage,
    messageHistory: [],
  });
  console.log(`Deep Mode - User: ${userMessage}`);
  console.log(`Deep Mode - Response length: ${response.message.length} chars`);
  expect(response.role).toEqual('assistant');
  expect(response.message).toBeTruthy();
  expect(response.message.length).toBeGreaterThan(100);
 });
 tap.test('Extended Thinking: should stop deep mode provider', async () => {
  await anthropicProviderDeep.stop();
 });
 // Test 'off' mode
 tap.test('Extended Thinking: should create Anthropic provider with thinking disabled', async () => {
  anthropicProviderOff = new smartai.AnthropicProvider({
    anthropicToken: await testQenv.getEnvVarOnDemand('ANTHROPIC_TOKEN'),
    extendedThinking: 'off',
  });
  await anthropicProviderOff.start();
  expect(anthropicProviderOff).toBeInstanceOf(smartai.AnthropicProvider);
 });
 tap.test('Extended Thinking: should chat with thinking disabled', async () => {
  const userMessage = 'What is 2 + 2?';
  const response = await anthropicProviderOff.chat({
    systemMessage: 'You are a helpful assistant.',
    userMessage: userMessage,
    messageHistory: [],
  });
  console.log(`Thinking Off - User: ${userMessage}`);
  console.log(`Thinking Off - Response: ${response.message}`);
  expect(response.role).toEqual('assistant');
  expect(response.message).toBeTruthy();
  expect(response.message).toInclude('4');
 });
 tap.test('Extended Thinking: should stop off mode provider', async () => {
  await anthropicProviderOff.stop();
 });
 // Test with vision method
 tap.test('Extended Thinking: should work with vision method', async () => {
  const provider = new smartai.AnthropicProvider({
    anthropicToken: await testQenv.getEnvVarOnDemand('ANTHROPIC_TOKEN'),
    extendedThinking: 'normal',
  });
  await provider.start();
  // Create a simple test image (1x1 red pixel PNG)
  const redPixelPng = Buffer.from(
    'iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mP8z8DwHwAFBQIAX8jx0gAAAABJRU5ErkJggg==',
    'base64'
  );
  const response = await provider.vision({
    image: redPixelPng,
    prompt: 'What color is this image?',
  });
  console.log(`Vision with Thinking - Response: ${response}`);
  expect(response).toBeTruthy();
  expect(response.toLowerCase()).toInclude('red');
  await provider.stop();
 });
 export default tap.start();
--- a/test/test.vision.anthropic.ts
+++ b/test/test.vision.anthropic.ts
@@ -1,8 +1,9 @@
-import { expect, tap } from '@push.rocks/tapbundle';
+import { expect, tap } from '@git.zone/tstest/tapbundle';
 import * as qenv from '@push.rocks/qenv';
-import * as smartfile from '@push.rocks/smartfile';
+import { SmartFs, SmartFsProviderNode } from '@push.rocks/smartfs';
 const testQenv = new qenv.Qenv('./', './.nogit/');
 const smartfs = new SmartFs(new SmartFsProviderNode());
 import * as smartai from '../ts/index.js';
@@ -21,7 +22,7 @@ tap.test('Anthropic Vision: should analyze coffee image with latte art', async (
  const imagePath = './test/testimages/coffee-dani/coffee.jpg';
  console.log(`Loading coffee image from: ${imagePath}`);
-  const imageBuffer = await smartfile.fs.toBuffer(imagePath);
+  const imageBuffer = await smartfs.file(imagePath).read();
  console.log(`Image loaded, size: ${imageBuffer.length} bytes`);
  const result = await anthropicProvider.vision({
@@ -45,7 +46,7 @@ tap.test('Anthropic Vision: should analyze laptop/workspace image', async () =>
  const imagePath = './test/testimages/laptop-nicolas/laptop.jpg';
  console.log(`Loading laptop image from: ${imagePath}`);
-  const imageBuffer = await smartfile.fs.toBuffer(imagePath);
+  const imageBuffer = await smartfs.file(imagePath).read();
  console.log(`Image loaded, size: ${imageBuffer.length} bytes`);
  const result = await anthropicProvider.vision({
@@ -69,7 +70,7 @@ tap.test('Anthropic Vision: should analyze receipt/document image', async () =>
  const imagePath = './test/testimages/receipt-annie/receipt.jpg';
  console.log(`Loading receipt image from: ${imagePath}`);
-  const imageBuffer = await smartfile.fs.toBuffer(imagePath);
+  const imageBuffer = await smartfs.file(imagePath).read();
  console.log(`Image loaded, size: ${imageBuffer.length} bytes`);
  const result = await anthropicProvider.vision({
--- a/ts/00_commitinfo_data.ts
+++ b/ts/00_commitinfo_data.ts
@@ -3,6 +3,6 @@
 */
 export const commitinfo = {
  name: '@push.rocks/smartai',
-  version: '0.7.7',
+  version: '0.12.0',
  description: 'SmartAi is a versatile TypeScript library designed to facilitate integration and interaction with various AI models, offering functionalities for chat, audio generation, document processing, and vision tasks.'
 }
--- a/ts/abstract.classes.multimodal.ts
+++ b/ts/abstract.classes.multimodal.ts
@@ -6,6 +6,8 @@ import * as plugins from './plugins.js';
 export interface ChatMessage {
  role: 'assistant' | 'user' | 'system';
  content: string;
  /** Base64-encoded images for vision-capable models */
  images?: string[];
 }
 /**
@@ -15,6 +17,16 @@ export interface ChatOptions {
  systemMessage: string;
  userMessage: string;
  messageHistory: ChatMessage[];
  /** Base64-encoded images for the current message (vision-capable models) */
  images?: string[];
 }
 /**
 * Options for streaming chat interactions
 */
 export interface StreamingChatOptions extends ChatOptions {
  /** Callback fired for each token during generation */
  onToken?: (token: string) => void;
 }
 /**
@@ -161,6 +173,14 @@ export abstract class MultiModalModel {
   */
  public abstract chatStream(input: ReadableStream<Uint8Array>): Promise<ReadableStream<string>>;
  /**
   * Streaming chat with token callback
   * Calls onToken for each token generated, returns final response
   * @param optionsArg Options containing system message, user message, message history, and onToken callback
   * @returns Promise resolving to the assistant's response
   */
  public chatStreaming?(optionsArg: StreamingChatOptions): Promise<ChatResponse>;
  /**
   * Text-to-speech conversion
   * @param optionsArg Options containing the message to convert to speech
--- a/ts/classes.conversation.ts
+++ b/ts/classes.conversation.ts
@@ -84,6 +84,18 @@ export class Conversation {
    return conversation;
  }
  public static async createWithMistral(smartaiRefArg: SmartAi) {
    if (!smartaiRefArg.mistralProvider) {
      throw new Error('Mistral provider not available');
    }
    const conversation = new Conversation(smartaiRefArg, {
      processFunction: async (input) => {
        return '' // TODO implement proper streaming
      }
    });
    return conversation;
  }
  public static async createWithXai(smartaiRefArg: SmartAi) {
    if (!smartaiRefArg.xaiProvider) {
      throw new Error('XAI provider not available');
--- a/ts/classes.smartai.ts
+++ b/ts/classes.smartai.ts
@@ -2,6 +2,7 @@ import { Conversation } from './classes.conversation.js';
 import * as plugins from './plugins.js';
 import { AnthropicProvider } from './provider.anthropic.js';
 import { ElevenLabsProvider } from './provider.elevenlabs.js';
 import { MistralProvider } from './provider.mistral.js';
 import { OllamaProvider } from './provider.ollama.js';
 import { OpenAiProvider } from './provider.openai.js';
 import { PerplexityProvider } from './provider.perplexity.js';
@@ -15,16 +16,33 @@ export interface ISmartAiOptions {
  anthropicToken?: string;
  perplexityToken?: string;
  groqToken?: string;
  mistralToken?: string;
  xaiToken?: string;
  elevenlabsToken?: string;
  exo?: {
    baseUrl?: string;
    apiKey?: string;
  };
  mistral?: {
    chatModel?: string;
    ocrModel?: string;
    tableFormat?: 'markdown' | 'html';
  };
  ollama?: {
    baseUrl?: string;
    model?: string;
    visionModel?: string;
    defaultOptions?: {
      num_ctx?: number;
      temperature?: number;
      top_k?: number;
      top_p?: number;
      repeat_penalty?: number;
      num_predict?: number;
      stop?: string[];
      seed?: number;
    };
    defaultTimeout?: number;
  };
  elevenlabs?: {
    defaultVoiceId?: string;
@@ -32,7 +50,7 @@ export interface ISmartAiOptions {
  };
 }
-export type TProvider = 'openai' | 'anthropic' | 'perplexity' | 'ollama' | 'exo' | 'groq' | 'xai' | 'elevenlabs';
+export type TProvider = 'openai' | 'anthropic' | 'perplexity' | 'ollama' | 'exo' | 'groq' | 'mistral' | 'xai' | 'elevenlabs';
 export class SmartAi {
  public options: ISmartAiOptions;
@@ -43,6 +61,7 @@ export class SmartAi {
  public ollamaProvider: OllamaProvider;
  public exoProvider: ExoProvider;
  public groqProvider: GroqProvider;
  public mistralProvider: MistralProvider;
  public xaiProvider: XAIProvider;
  public elevenlabsProvider: ElevenLabsProvider;
@@ -75,6 +94,15 @@ export class SmartAi {
      });
      await this.groqProvider.start();
    }
    if (this.options.mistralToken) {
      this.mistralProvider = new MistralProvider({
        mistralToken: this.options.mistralToken,
        chatModel: this.options.mistral?.chatModel,
        ocrModel: this.options.mistral?.ocrModel,
        tableFormat: this.options.mistral?.tableFormat,
      });
      await this.mistralProvider.start();
    }
    if (this.options.xaiToken) {
      this.xaiProvider = new XAIProvider({
        xaiToken: this.options.xaiToken,
@@ -94,6 +122,8 @@ export class SmartAi {
        baseUrl: this.options.ollama.baseUrl,
        model: this.options.ollama.model,
        visionModel: this.options.ollama.visionModel,
        defaultOptions: this.options.ollama.defaultOptions,
        defaultTimeout: this.options.ollama.defaultTimeout,
      });
      await this.ollamaProvider.start();
    }
@@ -119,6 +149,9 @@ export class SmartAi {
    if (this.groqProvider) {
      await this.groqProvider.stop();
    }
    if (this.mistralProvider) {
      await this.mistralProvider.stop();
    }
    if (this.xaiProvider) {
      await this.xaiProvider.stop();
    }
@@ -150,6 +183,8 @@ export class SmartAi {
        return Conversation.createWithOllama(this);
      case 'groq':
        return Conversation.createWithGroq(this);
      case 'mistral':
        return Conversation.createWithMistral(this);
      case 'xai':
        return Conversation.createWithXai(this);
      case 'elevenlabs':
--- a/ts/index.ts
+++ b/ts/index.ts
@@ -4,6 +4,7 @@ export * from './provider.openai.js';
 export * from './provider.anthropic.js';
 export * from './provider.perplexity.js';
 export * from './provider.groq.js';
 export * from './provider.mistral.js';
 export * from './provider.ollama.js';
 export * from './provider.xai.js';
 export * from './provider.exo.js';
--- a/ts/plugins.ts
+++ b/ts/plugins.ts
@@ -8,7 +8,7 @@ export {
 // @push.rocks scope
 import * as qenv from '@push.rocks/qenv';
 import * as smartarray from '@push.rocks/smartarray';
-import * as smartfile from '@push.rocks/smartfile';
+import * as smartfs from '@push.rocks/smartfs';
 import * as smartpath from '@push.rocks/smartpath';
 import * as smartpdf from '@push.rocks/smartpdf';
 import * as smartpromise from '@push.rocks/smartpromise';
@@ -18,7 +18,7 @@ import * as webstream from '@push.rocks/webstream';
 export {
  smartarray,
  qenv,
-  smartfile,
+  smartfs,
  smartpath,
  smartpdf,
  smartpromise,
@@ -28,9 +28,11 @@ export {
 // third party
 import * as anthropic from '@anthropic-ai/sdk';
 import * as mistralai from '@mistralai/mistralai';
 import * as openai from 'openai';
 export {
  anthropic,
  mistralai,
  openai,
 }
--- a/ts/provider.anthropic.ts
+++ b/ts/provider.anthropic.ts
@@ -20,6 +20,7 @@ export interface IAnthropicProviderOptions {
  enableWebSearch?: boolean;
  searchDomainAllowList?: string[];
  searchDomainBlockList?: string[];
  extendedThinking?: 'quick' | 'normal' | 'deep' | 'off';
 }
 export class AnthropicProvider extends MultiModalModel {
@@ -42,6 +43,25 @@ export class AnthropicProvider extends MultiModalModel {
    await super.stop();
  }
  /**
   * Returns the thinking configuration based on provider options.
   * Defaults to 'normal' mode (8000 tokens) if not specified.
   */
  private getThinkingConfig(): { type: 'enabled'; budget_tokens: number } | undefined {
    const mode = this.options.extendedThinking ?? 'normal';
    const budgetMap = {
      quick: 2048,
      normal: 8000,
      deep: 16000,
      off: 0,
    };
    const budget = budgetMap[mode];
    return budget > 0 ? { type: 'enabled', budget_tokens: budget } : undefined;
  }
  public async chatStream(input: ReadableStream<Uint8Array>): Promise<ReadableStream<string>> {
    // Create a TextDecoder to handle incoming chunks
    const decoder = new TextDecoder();
@@ -76,12 +96,14 @@ export class AnthropicProvider extends MultiModalModel {
        // If we have a complete message, send it to Anthropic
        if (currentMessage) {
          const thinkingConfig = this.getThinkingConfig();
          const stream = await this.anthropicApiClient.messages.create({
            model: 'claude-sonnet-4-5-20250929',
            messages: [{ role: currentMessage.role, content: currentMessage.content }],
            system: '',
            stream: true,
-            max_tokens: 4000,
+            max_tokens: 20000,
            ...(thinkingConfig && { thinking: thinkingConfig }),
          });
          // Process each chunk from Anthropic
@@ -120,6 +142,7 @@ export class AnthropicProvider extends MultiModalModel {
      content: msg.content
    }));
    const thinkingConfig = this.getThinkingConfig();
    const result = await this.anthropicApiClient.messages.create({
      model: 'claude-sonnet-4-5-20250929',
      system: optionsArg.systemMessage,
@@ -127,7 +150,8 @@ export class AnthropicProvider extends MultiModalModel {
        ...messages,
        { role: 'user' as const, content: optionsArg.userMessage }
      ],
-      max_tokens: 4000,
+      max_tokens: 20000,
      ...(thinkingConfig && { thinking: thinkingConfig }),
    });
    // Extract text content from the response
@@ -167,13 +191,15 @@ export class AnthropicProvider extends MultiModalModel {
      }
    ];
    const thinkingConfig = this.getThinkingConfig();
    const result = await this.anthropicApiClient.messages.create({
      model: 'claude-sonnet-4-5-20250929',
      messages: [{
        role: 'user',
        content
      }],
-      max_tokens: 1024
+      max_tokens: 10000,
      ...(thinkingConfig && { thinking: thinkingConfig }),
    });
    // Extract text content from the response
@@ -229,6 +255,7 @@ export class AnthropicProvider extends MultiModalModel {
      });
    }
    const thinkingConfig = this.getThinkingConfig();
    const result = await this.anthropicApiClient.messages.create({
      model: 'claude-sonnet-4-5-20250929',
      system: optionsArg.systemMessage,
@@ -236,7 +263,8 @@ export class AnthropicProvider extends MultiModalModel {
        ...messages,
        { role: 'user', content }
      ],
-      max_tokens: 4096
+      max_tokens: 20000,
      ...(thinkingConfig && { thinking: thinkingConfig }),
    });
    // Extract text content from the response
@@ -286,10 +314,14 @@ export class AnthropicProvider extends MultiModalModel {
      }
      // Configure the request based on search depth
-      const maxTokens = optionsArg.searchDepth === 'deep' ? 8192 :
+      const maxTokens = optionsArg.searchDepth === 'deep' ? 20000 :
-                        optionsArg.searchDepth === 'advanced' ? 6144 : 4096;
+                        optionsArg.searchDepth === 'advanced' ? 20000 : 20000;
      // Add thinking configuration if enabled
      const thinkingConfig = this.getThinkingConfig();
      // Create the research request
      // Note: When thinking is enabled, temperature must be 1 (or omitted)
      const requestParams: any = {
        model: 'claude-sonnet-4-5-20250929',
        system: systemMessage,
@@ -300,7 +332,8 @@ export class AnthropicProvider extends MultiModalModel {
          }
        ],
        max_tokens: maxTokens,
-        temperature: 0.7
+        // Only set temperature when thinking is NOT enabled
        ...(thinkingConfig ? {} : { temperature: 0.7 })
      };
      // Add tools if web search is enabled
@@ -308,6 +341,11 @@ export class AnthropicProvider extends MultiModalModel {
        requestParams.tools = tools;
      }
      // Add thinking configuration if enabled
      if (thinkingConfig) {
        requestParams.thinking = thinkingConfig;
      }
      // Execute the research request
      const result = await this.anthropicApiClient.messages.create(requestParams);
--- a/ts/provider.elevenlabs.ts
+++ b/ts/provider.elevenlabs.ts
@@ -1,4 +1,5 @@
 import * as plugins from './plugins.js';
 import { Readable } from 'stream';
 import { MultiModalModel } from './abstract.classes.multimodal.js';
 import type {
@@ -83,7 +84,8 @@ export class ElevenLabsProvider extends MultiModalModel {
      throw new Error(`ElevenLabs API error: ${response.status} ${response.statusText} - ${errorText}`);
    }
-    const nodeStream = response.streamNode();
+    const webStream = response.stream();
    const nodeStream = Readable.fromWeb(webStream as any);
    return nodeStream;
  }
--- a/ts/provider.mistral.ts
+++ b/ts/provider.mistral.ts
@@ -0,0 +1,352 @@
 import * as plugins from './plugins.js';
 import { MultiModalModel } from './abstract.classes.multimodal.js';
 import type {
  ChatOptions,
  ChatResponse,
  ChatMessage,
  ResearchOptions,
  ResearchResponse,
  ImageGenerateOptions,
  ImageEditOptions,
  ImageResponse
 } from './abstract.classes.multimodal.js';
 export interface IMistralProviderOptions {
  mistralToken: string;
  chatModel?: string;       // default: 'mistral-large-latest'
  ocrModel?: string;        // default: 'mistral-ocr-latest'
  tableFormat?: 'markdown' | 'html';
 }
 export class MistralProvider extends MultiModalModel {
  private options: IMistralProviderOptions;
  public mistralClient: plugins.mistralai.Mistral;
  constructor(optionsArg: IMistralProviderOptions) {
    super();
    this.options = optionsArg;
  }
  async start() {
    await super.start();
    this.mistralClient = new plugins.mistralai.Mistral({
      apiKey: this.options.mistralToken,
    });
  }
  async stop() {
    await super.stop();
  }
  /**
   * Synchronous chat interaction using Mistral's chat API
   */
  public async chat(optionsArg: ChatOptions): Promise<ChatResponse> {
    // Convert message history to Mistral format
    const messages: Array<{
      role: 'system' | 'user' | 'assistant';
      content: string;
    }> = [];
    // Add system message first
    if (optionsArg.systemMessage) {
      messages.push({
        role: 'system',
        content: optionsArg.systemMessage
      });
    }
    // Add message history
    for (const msg of optionsArg.messageHistory) {
      messages.push({
        role: msg.role === 'system' ? 'system' : msg.role === 'assistant' ? 'assistant' : 'user',
        content: msg.content
      });
    }
    // Add current user message
    messages.push({
      role: 'user',
      content: optionsArg.userMessage
    });
    const result = await this.mistralClient.chat.complete({
      model: this.options.chatModel || 'mistral-large-latest',
      messages: messages,
    });
    // Extract content from response
    const choice = result.choices?.[0];
    let content = '';
    if (choice?.message?.content) {
      if (typeof choice.message.content === 'string') {
        content = choice.message.content;
      } else if (Array.isArray(choice.message.content)) {
        // Handle array of content chunks
        content = choice.message.content
          .map((chunk: any) => {
            if (typeof chunk === 'string') return chunk;
            if (chunk && typeof chunk === 'object' && 'text' in chunk) return chunk.text;
            return '';
          })
          .join('');
      }
    }
    return {
      role: 'assistant',
      message: content,
    };
  }
  /**
   * Streaming chat using Mistral's streaming API
   */
  public async chatStream(input: ReadableStream<Uint8Array>): Promise<ReadableStream<string>> {
    const decoder = new TextDecoder();
    let buffer = '';
    const mistralClient = this.mistralClient;
    const chatModel = this.options.chatModel || 'mistral-large-latest';
    const transform = new TransformStream<Uint8Array, string>({
      async transform(chunk, controller) {
        buffer += decoder.decode(chunk, { stream: true });
        // Try to parse complete JSON messages from the buffer
        while (true) {
          const newlineIndex = buffer.indexOf('\n');
          if (newlineIndex === -1) break;
          const line = buffer.slice(0, newlineIndex);
          buffer = buffer.slice(newlineIndex + 1);
          if (line.trim()) {
            try {
              const message = JSON.parse(line);
              // Build messages array
              const messages: Array<{
                role: 'system' | 'user' | 'assistant';
                content: string;
              }> = [];
              if (message.systemMessage) {
                messages.push({
                  role: 'system',
                  content: message.systemMessage
                });
              }
              messages.push({
                role: message.role === 'assistant' ? 'assistant' : 'user',
                content: message.content
              });
              // Use Mistral streaming
              const stream = await mistralClient.chat.stream({
                model: chatModel,
                messages: messages,
              });
              // Process streaming events
              for await (const event of stream) {
                const delta = event.data?.choices?.[0]?.delta;
                if (delta?.content) {
                  if (typeof delta.content === 'string') {
                    controller.enqueue(delta.content);
                  } else if (Array.isArray(delta.content)) {
                    for (const chunk of delta.content) {
                      if (typeof chunk === 'string') {
                        controller.enqueue(chunk);
                      } else if (chunk && typeof chunk === 'object' && 'text' in chunk) {
                        controller.enqueue((chunk as any).text);
                      }
                    }
                  }
                }
              }
            } catch (e) {
              console.error('Failed to parse message:', e);
            }
          }
        }
      },
      flush(controller) {
        if (buffer.trim()) {
          try {
            const message = JSON.parse(buffer);
            controller.enqueue(message.content || '');
          } catch (e) {
            console.error('Failed to parse remaining buffer:', e);
          }
        }
      }
    });
    return input.pipeThrough(transform);
  }
  /**
   * Audio generation is not supported by Mistral
   */
  public async audio(optionsArg: { message: string }): Promise<NodeJS.ReadableStream> {
    throw new Error('Audio generation is not supported by Mistral. Please use ElevenLabs or OpenAI provider for audio generation.');
  }
  /**
   * Vision using Mistral's OCR API for image analysis
   */
  public async vision(optionsArg: { image: Buffer; prompt: string }): Promise<string> {
    const base64Image = optionsArg.image.toString('base64');
    // Detect image type from buffer header
    let mimeType = 'image/jpeg';
    if (optionsArg.image[0] === 0x89 && optionsArg.image[1] === 0x50) {
      mimeType = 'image/png';
    } else if (optionsArg.image[0] === 0x47 && optionsArg.image[1] === 0x49) {
      mimeType = 'image/gif';
    } else if (optionsArg.image[0] === 0x52 && optionsArg.image[1] === 0x49) {
      mimeType = 'image/webp';
    }
    // Use OCR API with image data URL
    const ocrResult = await this.mistralClient.ocr.process({
      model: this.options.ocrModel || 'mistral-ocr-latest',
      document: {
        imageUrl: `data:${mimeType};base64,${base64Image}`,
        type: 'image_url',
      },
    });
    // Combine markdown from all pages
    const extractedText = ocrResult.pages.map(page => page.markdown).join('\n\n');
    // If a prompt is provided, use chat to analyze the extracted text
    if (optionsArg.prompt && optionsArg.prompt.trim()) {
      const chatResponse = await this.chat({
        systemMessage: 'You are an assistant analyzing image content. The following is text extracted from an image using OCR.',
        userMessage: `${optionsArg.prompt}\n\nExtracted content:\n${extractedText}`,
        messageHistory: [],
      });
      return chatResponse.message;
    }
    return extractedText;
  }
  /**
   * Document processing using Mistral's OCR API
   * PDFs are uploaded via Files API first, then processed with OCR
   */
  public async document(optionsArg: {
    systemMessage: string;
    userMessage: string;
    pdfDocuments: Uint8Array[];
    messageHistory: ChatMessage[];
  }): Promise<{ message: any }> {
    const extractedTexts: string[] = [];
    const uploadedFileIds: string[] = [];
    try {
      // Process each PDF document using Mistral OCR
      for (let i = 0; i < optionsArg.pdfDocuments.length; i++) {
        const pdfDocument = optionsArg.pdfDocuments[i];
        // Upload the PDF to Mistral's Files API first
        const uploadResult = await this.mistralClient.files.upload({
          file: {
            fileName: `document_${i + 1}.pdf`,
            content: pdfDocument,
          },
          purpose: 'ocr',
        });
        uploadedFileIds.push(uploadResult.id);
        // Now use OCR with the uploaded file
        const ocrResult = await this.mistralClient.ocr.process({
          model: this.options.ocrModel || 'mistral-ocr-latest',
          document: {
            type: 'file',
            fileId: uploadResult.id,
          },
          tableFormat: this.options.tableFormat || 'markdown',
        });
        // Combine all page markdown with page separators
        const pageTexts = ocrResult.pages.map((page, index) => {
          let pageContent = `--- Page ${index + 1} ---\n${page.markdown}`;
          // Include tables if present
          if (page.tables && page.tables.length > 0) {
            pageContent += '\n\n**Tables:**\n' + page.tables.map((t: any) => t.markdown || t.html || '').join('\n');
          }
          // Include header/footer if present
          if (page.header) {
            pageContent = `Header: ${page.header}\n${pageContent}`;
          }
          if (page.footer) {
            pageContent += `\nFooter: ${page.footer}`;
          }
          return pageContent;
        }).join('\n\n');
        extractedTexts.push(pageTexts);
      }
      // Combine all document texts
      const allDocumentText = extractedTexts.length === 1
        ? extractedTexts[0]
        : extractedTexts.map((text, i) => `=== Document ${i + 1} ===\n${text}`).join('\n\n');
      // Use chat API to process the extracted text with the user's query
      const chatResponse = await this.chat({
        systemMessage: optionsArg.systemMessage || 'You are a helpful assistant analyzing document content.',
        userMessage: `${optionsArg.userMessage}\n\n---\nDocument Content:\n${allDocumentText}`,
        messageHistory: optionsArg.messageHistory,
      });
      return {
        message: {
          role: 'assistant',
          content: chatResponse.message
        }
      };
    } finally {
      // Clean up uploaded files
      for (const fileId of uploadedFileIds) {
        try {
          await this.mistralClient.files.delete({ fileId });
        } catch (cleanupError) {
          // Ignore cleanup errors - files may have already been auto-deleted
          console.warn(`Failed to delete temporary file ${fileId}:`, cleanupError);
        }
      }
    }
  }
  /**
   * Research is not natively supported by Mistral
   */
  public async research(optionsArg: ResearchOptions): Promise<ResearchResponse> {
    throw new Error('Research/web search is not supported by Mistral. Please use Perplexity or Anthropic provider for research capabilities.');
  }
  /**
   * Image generation is not supported by Mistral
   */
  public async imageGenerate(optionsArg: ImageGenerateOptions): Promise<ImageResponse> {
    throw new Error('Image generation is not supported by Mistral. Please use OpenAI provider for image generation.');
  }
  /**
   * Image editing is not supported by Mistral
   */
  public async imageEdit(optionsArg: ImageEditOptions): Promise<ImageResponse> {
    throw new Error('Image editing is not supported by Mistral. Please use OpenAI provider for image editing.');
  }
 }
--- a/ts/provider.ollama.ts
+++ b/ts/provider.ollama.ts
@@ -9,13 +9,65 @@ import type {
  ResearchResponse,
  ImageGenerateOptions,
  ImageEditOptions,
-  ImageResponse
+  ImageResponse,
  StreamingChatOptions
 } from './abstract.classes.multimodal.js';
 /**
 * Ollama model runtime options
 * @see https://github.com/ollama/ollama/blob/main/docs/modelfile.md
 */
 export interface IOllamaModelOptions {
  num_ctx?: number;       // Context window (default: 2048)
  temperature?: number;   // 0 = deterministic (default: 0.8)
  top_k?: number;         // Top-k sampling (default: 40)
  top_p?: number;         // Nucleus sampling (default: 0.9)
  repeat_penalty?: number;// Repeat penalty (default: 1.1)
  num_predict?: number;   // Max tokens to predict
  stop?: string[];        // Stop sequences
  seed?: number;          // Random seed for reproducibility
 }
 export interface IOllamaProviderOptions {
  baseUrl?: string;
  model?: string;
  visionModel?: string; // Model to use for vision tasks (e.g. 'llava')
  defaultOptions?: IOllamaModelOptions;  // Default model options
  defaultTimeout?: number;               // Default timeout in ms (default: 120000)
 }
 /**
 * Extended chat options with Ollama-specific settings
 */
 export interface IOllamaChatOptions extends ChatOptions {
  options?: IOllamaModelOptions;  // Per-request model options
  timeout?: number;               // Per-request timeout in ms
  model?: string;                 // Per-request model override
  // images is inherited from ChatOptions
 }
 /**
 * Chunk emitted during streaming
 */
 export interface IOllamaStreamChunk {
  content: string;
  thinking?: string;  // For models with extended thinking
  done: boolean;
  stats?: {
    totalDuration?: number;
    evalCount?: number;
  };
 }
 /**
 * Extended chat response with Ollama-specific fields
 */
 export interface IOllamaChatResponse extends ChatResponse {
  thinking?: string;
  stats?: {
    totalDuration?: number;
    evalCount?: number;
  };
 }
 export class OllamaProvider extends MultiModalModel {
@@ -23,6 +75,8 @@ export class OllamaProvider extends MultiModalModel {
  private baseUrl: string;
  private model: string;
  private visionModel: string;
  private defaultOptions: IOllamaModelOptions;
  private defaultTimeout: number;
  constructor(optionsArg: IOllamaProviderOptions = {}) {
    super();
@@ -30,6 +84,8 @@ export class OllamaProvider extends MultiModalModel {
    this.baseUrl = optionsArg.baseUrl || 'http://localhost:11434';
    this.model = optionsArg.model || 'llama2';
    this.visionModel = optionsArg.visionModel || 'llava';
    this.defaultOptions = optionsArg.defaultOptions || {};
    this.defaultTimeout = optionsArg.defaultTimeout || 120000;
  }
  async start() {
@@ -148,13 +204,33 @@ export class OllamaProvider extends MultiModalModel {
  // Implementing the synchronous chat interaction
  public async chat(optionsArg: ChatOptions): Promise<ChatResponse> {
    // Format messages for Ollama
    const historyMessages = optionsArg.messageHistory.map((msg) => {
      const formatted: { role: string; content: string; images?: string[] } = {
        role: msg.role,
        content: msg.content,
      };
      if (msg.images && msg.images.length > 0) {
        formatted.images = msg.images;
      }
      return formatted;
    });
    // Build user message with optional images
    const userMessage: { role: string; content: string; images?: string[] } = {
      role: 'user',
      content: optionsArg.userMessage,
    };
    if (optionsArg.images && optionsArg.images.length > 0) {
      userMessage.images = optionsArg.images;
    }
    const messages = [
      { role: 'system', content: optionsArg.systemMessage },
-      ...optionsArg.messageHistory,
+      ...historyMessages,
-      { role: 'user', content: optionsArg.userMessage }
+      userMessage,
    ];
-    // Make API call to Ollama
+    // Make API call to Ollama with defaultOptions and timeout
    const response = await fetch(`${this.baseUrl}/api/chat`, {
      method: 'POST',
      headers: {
@@ -163,8 +239,10 @@ export class OllamaProvider extends MultiModalModel {
      body: JSON.stringify({
        model: this.model,
        messages: messages,
-        stream: false
+        stream: false,
        options: this.defaultOptions,
      }),
      signal: AbortSignal.timeout(this.defaultTimeout),
    });
    if (!response.ok) {
@@ -179,6 +257,214 @@ export class OllamaProvider extends MultiModalModel {
    };
  }
  /**
   * Streaming chat with token callback (implements MultiModalModel interface)
   * Calls onToken for each token generated during the response
   */
  public async chatStreaming(optionsArg: StreamingChatOptions): Promise<ChatResponse> {
    const onToken = optionsArg.onToken;
    // Use existing collectStreamResponse with callback, including images
    const response = await this.collectStreamResponse(
      {
        systemMessage: optionsArg.systemMessage,
        userMessage: optionsArg.userMessage,
        messageHistory: optionsArg.messageHistory,
        images: optionsArg.images,
      },
      (chunk) => {
        if (onToken) {
          if (chunk.thinking) onToken(chunk.thinking);
          if (chunk.content) onToken(chunk.content);
        }
      }
    );
    return {
      role: 'assistant' as const,
      message: response.message,
    };
  }
  /**
   * Streaming chat with async iteration and options support
   */
  public async chatStreamResponse(
    optionsArg: IOllamaChatOptions
  ): Promise<AsyncIterable<IOllamaStreamChunk>> {
    const model = optionsArg.model || this.model;
    const timeout = optionsArg.timeout || this.defaultTimeout;
    const modelOptions = { ...this.defaultOptions, ...optionsArg.options };
    // Format history messages with optional images
    const historyMessages = optionsArg.messageHistory.map((msg) => {
      const formatted: { role: string; content: string; images?: string[] } = {
        role: msg.role,
        content: msg.content,
      };
      if (msg.images && msg.images.length > 0) {
        formatted.images = msg.images;
      }
      return formatted;
    });
    // Build user message with optional images
    const userMessage: { role: string; content: string; images?: string[] } = {
      role: 'user',
      content: optionsArg.userMessage,
    };
    if (optionsArg.images && optionsArg.images.length > 0) {
      userMessage.images = optionsArg.images;
    }
    const messages = [
      { role: 'system', content: optionsArg.systemMessage },
      ...historyMessages,
      userMessage,
    ];
    const response = await fetch(`${this.baseUrl}/api/chat`, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify({
        model,
        messages,
        stream: true,
        options: modelOptions,
      }),
      signal: AbortSignal.timeout(timeout),
    });
    if (!response.ok) {
      throw new Error(`Ollama API error: ${response.status}`);
    }
    const reader = response.body!.getReader();
    const decoder = new TextDecoder();
    return {
      [Symbol.asyncIterator]: async function* () {
        let buffer = '';
        try {
          while (true) {
            const { done, value } = await reader.read();
            if (done) break;
            buffer += decoder.decode(value, { stream: true });
            const lines = buffer.split('\n');
            buffer = lines.pop() || '';
            for (const line of lines) {
              if (!line.trim()) continue;
              try {
                const json = JSON.parse(line);
                yield {
                  content: json.message?.content || '',
                  thinking: json.message?.thinking,
                  done: json.done || false,
                  stats: json.done ? {
                    totalDuration: json.total_duration,
                    evalCount: json.eval_count,
                  } : undefined,
                } as IOllamaStreamChunk;
              } catch { /* skip malformed */ }
            }
          }
        } finally {
          reader.releaseLock();
        }
      }
    };
  }
  /**
   * Stream and collect full response with optional progress callback
   */
  public async collectStreamResponse(
    optionsArg: IOllamaChatOptions,
    onChunk?: (chunk: IOllamaStreamChunk) => void
  ): Promise<IOllamaChatResponse> {
    const stream = await this.chatStreamResponse(optionsArg);
    let content = '';
    let thinking = '';
    let stats: IOllamaChatResponse['stats'];
    for await (const chunk of stream) {
      if (chunk.content) content += chunk.content;
      if (chunk.thinking) thinking += chunk.thinking;
      if (chunk.stats) stats = chunk.stats;
      if (onChunk) onChunk(chunk);
    }
    return {
      role: 'assistant' as const,
      message: content,
      thinking: thinking || undefined,
      stats,
    };
  }
  /**
   * Non-streaming chat with full options support
   */
  public async chatWithOptions(optionsArg: IOllamaChatOptions): Promise<IOllamaChatResponse> {
    const model = optionsArg.model || this.model;
    const timeout = optionsArg.timeout || this.defaultTimeout;
    const modelOptions = { ...this.defaultOptions, ...optionsArg.options };
    // Format history messages with optional images
    const historyMessages = optionsArg.messageHistory.map((msg) => {
      const formatted: { role: string; content: string; images?: string[] } = {
        role: msg.role,
        content: msg.content,
      };
      if (msg.images && msg.images.length > 0) {
        formatted.images = msg.images;
      }
      return formatted;
    });
    // Build user message with optional images
    const userMessage: { role: string; content: string; images?: string[] } = {
      role: 'user',
      content: optionsArg.userMessage,
    };
    if (optionsArg.images && optionsArg.images.length > 0) {
      userMessage.images = optionsArg.images;
    }
    const messages = [
      { role: 'system', content: optionsArg.systemMessage },
      ...historyMessages,
      userMessage,
    ];
    const response = await fetch(`${this.baseUrl}/api/chat`, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify({
        model,
        messages,
        stream: false,
        options: modelOptions,
      }),
      signal: AbortSignal.timeout(timeout),
    });
    if (!response.ok) {
      throw new Error(`Ollama API error: ${response.statusText}`);
    }
    const result = await response.json();
    return {
      role: 'assistant' as const,
      message: result.message.content,
      thinking: result.message.thinking,
      stats: {
        totalDuration: result.total_duration,
        evalCount: result.eval_count,
      },
    };
  }
  public async audio(optionsArg: { message: string }): Promise<NodeJS.ReadableStream> {
    throw new Error('Audio generation is not supported by Ollama.');
  }
--- a/ts/provider.openai.ts
+++ b/ts/provider.openai.ts
@@ -1,6 +1,7 @@
 import * as plugins from './plugins.js';
 import * as paths from './paths.js';
 import { Readable } from 'stream';
 import { toFile } from 'openai';
 // Custom type definition for chat completion messages
 export type TChatCompletionRequestMessage = { 
@@ -405,16 +406,19 @@ export class OpenAiProvider extends MultiModalModel {
    const model = optionsArg.model || this.options.imageModel || 'gpt-image-1';
    try {
      // Convert Buffer to uploadable file format for OpenAI API
      const imageFile = await toFile(optionsArg.image, 'image.png', { type: 'image/png' });
      const requestParams: any = {
        model,
-        image: optionsArg.image,
+        image: imageFile,
        prompt: optionsArg.prompt,
        n: optionsArg.n || 1,
      };
-      // Add mask if provided
+      // Add mask if provided (also convert to file format)
      if (optionsArg.mask) {
-        requestParams.mask = optionsArg.mask;
+        requestParams.mask = await toFile(optionsArg.mask, 'mask.png', { type: 'image/png' });
      }
      // Add gpt-image-1 specific parameters
--- a/tsconfig.json
+++ b/tsconfig.json
@@ -6,9 +6,9 @@
    "module": "NodeNext",
    "moduleResolution": "NodeNext",
    "esModuleInterop": true,
-    "verbatimModuleSyntax": true
+    "verbatimModuleSyntax": true,
    "baseUrl": ".",
    "paths": {}
  },
-  "exclude": [
+  "exclude": ["dist_*/**/*.d.ts"]
    "dist_*/**/*.d.ts"
  ]
 }
Author	SHA1	Message	Date
Juergen Kunz	ae8d3ccf33	v0.12.0 Some checks failed Default (tags) / security (push) Failing after 1s Details Default (tags) / test (push) Failing after 1s Details Default (tags) / release (push) Has been skipped Details Default (tags) / metadata (push) Has been skipped Details	2026-01-20 01:10:27 +00:00
Juergen Kunz	3b900d0ba9	feat(ollama): add support for base64-encoded images in chat messages and forward them to the Ollama provider	2026-01-20 01:10:27 +00:00
Juergen Kunz	d49152390f	v0.11.1	2026-01-20 00:37:59 +00:00
Juergen Kunz	d615ec9227	feat(streaming): add chatStreaming method with token callback for real-time generation progress - Add StreamingChatOptions interface with onToken callback - Add optional chatStreaming method to MultiModalModel abstract class - Implement chatStreaming in OllamaProvider using collectStreamResponse	2026-01-20 00:37:49 +00:00
Juergen Kunz	dfa863ee7d	v0.11.0 Some checks failed Default (tags) / security (push) Failing after 1s Details Default (tags) / test (push) Failing after 1s Details Default (tags) / release (push) Has been skipped Details Default (tags) / metadata (push) Has been skipped Details	2026-01-20 00:12:21 +00:00
Juergen Kunz	c84ede1f1d	feat(ollama): support defaultOptions and defaultTimeout for ollama provider	2026-01-20 00:12:21 +00:00
Juergen Kunz	4937dbf6ab	v0.10.1 Some checks failed Default (tags) / security (push) Failing after 1s Details Default (tags) / test (push) Failing after 1s Details Default (tags) / release (push) Has been skipped Details Default (tags) / metadata (push) Has been skipped Details	2026-01-20 00:03:06 +00:00
Juergen Kunz	8cb052449e	fix(): no changes detected — no release necessary	2026-01-20 00:03:06 +00:00
Juergen Kunz	126e9b239b	feat(OllamaProvider): add model options, streaming support, and thinking tokens - Add IOllamaModelOptions interface for runtime options (num_ctx, temperature, etc.) - Extend IOllamaProviderOptions with defaultOptions and defaultTimeout - Add IOllamaChatOptions for per-request overrides - Add IOllamaStreamChunk and IOllamaChatResponse interfaces - Add chatStreamResponse() for async iteration with options - Add collectStreamResponse() for streaming with progress callback - Add chatWithOptions() for non-streaming with full options - Update chat() to use defaultOptions and defaultTimeout	2026-01-20 00:02:45 +00:00
Juergen Kunz	a556053510	v0.10.0 Some checks failed Default (tags) / security (push) Failing after 1s Details Default (tags) / test (push) Failing after 1s Details Default (tags) / release (push) Has been skipped Details Default (tags) / metadata (push) Has been skipped Details	2026-01-18 22:11:52 +00:00
Juergen Kunz	e4dc81edc9	feat(mistral): add Mistral provider with native PDF OCR and chat integration	2026-01-18 22:11:52 +00:00
Juergen Kunz	6f79dc3535	v0.9.0 Some checks failed Default (tags) / security (push) Failing after 1s Details Default (tags) / test (push) Failing after 1s Details Default (tags) / release (push) Has been skipped Details Default (tags) / metadata (push) Has been skipped Details	2026-01-18 16:26:16 +00:00
Juergen Kunz	b4ced080f2	feat(providers): Add Anthropic extended thinking and adapt providers to new streaming/file APIs; bump dependencies and update docs, tests and configuration	2026-01-18 16:26:16 +00:00
Juergen Kunz	e8a2a3ff1b	0.8.0 Some checks failed Default (tags) / security (push) Failing after 24s Details Default (tags) / test (push) Failing after 14s Details Default (tags) / release (push) Has been skipped Details Default (tags) / metadata (push) Has been skipped Details	2025-10-30 12:11:18 +00:00
Juergen Kunz	cbc9d8d45b	feat(provider.anthropic): Add extended thinking modes to AnthropicProvider and apply thinking budgets to API calls	2025-10-30 12:11:18 +00:00