v0.11.0

feat(ollama): support defaultOptions and defaultTimeout for ollama provider
v0.10.1
2026-01-20 00:12:21 +00:00 · 2026-01-20 00:12:21 +00:00 · 2026-01-20 00:03:06 +00:00 · 2026-01-20 00:03:06 +00:00 · 2026-01-20 00:02:45 +00:00 · 2026-01-18 22:11:52 +00:00
19 changed files with 988 additions and 69 deletions
--- a/changelog.md
+++ b/changelog.md
@@ -1,5 +1,27 @@
 # Changelog

+## 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

--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
 {
  "name": "@push.rocks/smartai",
-  "version": "0.9.0",
+  "version": "0.11.0",
  "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",
@@ -18,21 +18,22 @@
    "@git.zone/tsbuild": "^4.1.2",
    "@git.zone/tsbundle": "^2.8.1",
    "@git.zone/tsrun": "^2.0.1",
-    "@git.zone/tstest": "^3.1.4",
+    "@git.zone/tstest": "^3.1.6",
    "@push.rocks/qenv": "^6.1.3",
-    "@types/node": "^22.15.17",
+    "@types/node": "^25.0.9",
    "typescript": "^5.9.3"
  },
  "dependencies": {
    "@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": "^5.0.1",
    "@push.rocks/webstream": "^1.0.10",
-    "openai": "^5.12.2"
+    "openai": "^6.16.0"
  },
  "repository": {
    "type": "git",
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
@@ -11,12 +11,15 @@ importers:
      '@anthropic-ai/sdk':
        specifier: ^0.71.2
        version: 0.71.2(zod@3.25.76)
+      '@mistralai/mistralai':
+        specifier: ^1.12.0
+        version: 1.12.0
      '@push.rocks/smartarray':
        specifier: ^1.1.0
        version: 1.1.0
-      '@push.rocks/smartfile':
-        specifier: ^11.2.7
-        version: 11.2.7
+      '@push.rocks/smartfs':
+        specifier: ^1.3.1
+        version: 1.3.1
      '@push.rocks/smartpath':
        specifier: ^6.0.0
        version: 6.0.0
@@ -33,8 +36,8 @@ importers:
        specifier: ^1.0.10
        version: 1.0.10
      openai:
-        specifier: ^5.12.2
-        version: 5.12.2(ws@8.18.3)(zod@3.25.76)
+        specifier: ^6.16.0
+        version: 6.16.0(ws@8.18.3)(zod@3.25.76)
    devDependencies:
      '@git.zone/tsbuild':
        specifier: ^4.1.2
@@ -46,14 +49,14 @@ importers:
        specifier: ^2.0.1
        version: 2.0.1
      '@git.zone/tstest':
-        specifier: ^3.1.4
-        version: 3.1.4(@aws-sdk/credential-providers@3.808.0)(socks@2.8.4)(typescript@5.9.3)
+        specifier: ^3.1.6
+        version: 3.1.6(@aws-sdk/credential-providers@3.808.0)(socks@2.8.4)(typescript@5.9.3)
      '@push.rocks/qenv':
        specifier: ^6.1.3
        version: 6.1.3
      '@types/node':
-        specifier: ^22.15.17
-        version: 22.15.17
+        specifier: ^25.0.9
+        version: 25.0.9
      typescript:
        specifier: ^5.9.3
        version: 5.9.3
@@ -813,8 +816,8 @@ packages:
    resolution: {integrity: sha512-NEcnsjvlC1o3Z6SS3VhKCf6Ev+Sh4EAinmggslrIR/ppMrvjDbXNFXoyr3PB+GLeSAR0JRZ1fGvVYjpEzjBdIg==}
    hasBin: true

-  '@git.zone/tstest@3.1.4':
-    resolution: {integrity: sha512-S7kubbb0yLYOh/QAzFsjG6a20lZiyNKo4pt0yK1yvd9I7X8Rw6/mCT/BicLkan7G7Nk7scUfxaK9+aFsHmdQdw==}
+  '@git.zone/tstest@3.1.6':
+    resolution: {integrity: sha512-xRGc6wO4rJ6mohPCMIBDRH+oNjiIvX6Jeo8v/Y5o5VyKSHFmqol7FCKSBrojMcqgBpESnLHFPJAAOmT9W3JV8Q==}
    hasBin: true

  '@happy-dom/global-registrator@15.11.7':
@@ -1132,6 +1135,9 @@ packages:
  '@lit/reactive-element@2.1.1':
    resolution: {integrity: sha512-N+dm5PAYdQ8e6UlywyyrgI2t++wFGXfHx+dSJ1oBrg6FAxUj40jId++EaRm80MKX5JnlH1sBsyZ5h0bcZKemCg==}

+  '@mistralai/mistralai@1.12.0':
+    resolution: {integrity: sha512-oDr1hcS3wsIT/QupBG93TNiA5kilwBYoAIyl5BNYqMM2Ix/xsNq+wT8b++uhp/GTUMx44n+8Bn1mkATbwxe6bQ==}
+
  '@mixmark-io/domino@2.2.0':
    resolution: {integrity: sha512-Y28PR25bHXUg88kCV7nivXrP2Nj2RueZ3/l/jdx6J9f8J4nsEGcgX0Qe6lt7Pa+J79+kPiJU3LguR6O/6zrLOw==}

@@ -2491,8 +2497,11 @@ packages:
  '@types/node@16.9.1':
    resolution: {integrity: sha512-QpLcX9ZSsq3YYUUnD3nFDY8H7wctAhQj/TFKL8Ya8v5fMm3CFXxo8zStsLAl780ltoYoo1WvKUVGBQK+1ifr7g==}

-  '@types/node@22.15.17':
-    resolution: {integrity: sha512-wIX2aSZL5FE+MR0JlvF87BNVrtFWf6AE6rxSE9X7OwnVvoyCQjpzSRJ+M87se/4QCkCiebQAqrJ0y6fwIyi7nw==}
+  '@types/node@22.19.7':
+    resolution: {integrity: sha512-MciR4AKGHWl7xwxkBa6xUGxQJ4VBOmPTF7sL+iGzuahOFaO0jHCsuEfS80pan1ef4gWId1oWOweIhrDEYLuaOw==}
+
+  '@types/node@25.0.9':
+    resolution: {integrity: sha512-/rpCXHlCWeqClNBwUhDcusJxXYDjZTyE8v5oTO7WbL8eij2nKhUeU89/6xgjU7N4/Vh3He0BtyhJdQbDyhiXAw==}

  '@types/ping@0.4.4':
    resolution: {integrity: sha512-ifvo6w2f5eJYlXm+HiVx67iJe8WZp87sfa683nlqED5Vnt9Z93onkokNoWqOG21EaE8fMxyKPobE+mkPEyxsdw==}
@@ -4019,12 +4028,12 @@ packages:
    resolution: {integrity: sha512-7x81NCL719oNbsq/3mh+hVrAWmFuEYUqrq/Iw3kUzH8ReypT9QQ0BLoJS7/G9k6N81XjW4qHWtjWwe/9eLy1EQ==}
    engines: {node: '>=12'}

-  openai@5.12.2:
-    resolution: {integrity: sha512-xqzHHQch5Tws5PcKR2xsZGX9xtch+JQFz5zb14dGqlshmmDAFBFEWmeIpf7wVqWV+w7Emj7jRgkNJakyKE0tYQ==}
+  openai@6.16.0:
+    resolution: {integrity: sha512-fZ1uBqjFUjXzbGc35fFtYKEOxd20kd9fDpFeqWtsOZWiubY8CZ1NAlXHW3iathaFvqmNtCWMIsosCuyeI7Joxg==}
    hasBin: true
    peerDependencies:
      ws: ^8.18.0
-      zod: ^3.23.8
+      zod: ^3.25 || ^4.0
    peerDependenciesMeta:
      ws:
        optional: true
@@ -4715,6 +4724,9 @@ packages:
  undici-types@6.21.0:
    resolution: {integrity: sha512-iwDZqg0QAGrg9Rav5H4n0M64c3mkR59cJ6wQp+7C4nI0gsmExaedaYLNO44eT4AtBBwjbTiGPMlt2Md0T9H9JQ==}

+  undici-types@7.16.0:
+    resolution: {integrity: sha512-Zz+aZWSj8LE6zoxD+xrjh4VfkIG8Ya6LvYkZqtUQGJPZjYl53ypCaUwWqo7eI0x66KBGeRo+mlBEkMSeSZ38Nw==}
+
  unified@11.0.5:
    resolution: {integrity: sha512-xKvGhPWw3k84Qjh8bI3ZeJjqnyadK+GEFtazSfZv/rKeTkTjOJho6mFqh2SM96iIcZokxiOpg78GazTSg8+KHA==}

@@ -4895,6 +4907,11 @@ packages:
    resolution: {integrity: sha512-U/PBtDf35ff0D8X8D0jfdzHYEPFxAI7jJlxZXwCSez5M3190m+QobIfh+sWDWSHMCWWJN2AWamkegn6vr6YBTw==}
    engines: {node: '>=18'}

+  zod-to-json-schema@3.25.1:
+    resolution: {integrity: sha512-pM/SU9d3YAggzi6MtR4h7ruuQlqKtad8e9S0fmxcMi+ueAK5Korys/aWcV9LIIHTVbj01NdzxcnXSN+O74ZIVA==}
+    peerDependencies:
+      zod: ^3.25 || ^4
+
  zod@3.24.2:
    resolution: {integrity: sha512-lY7CDW43ECgW9u1TcT3IoXHflywfVqDYze4waEz812jR/bZ8FHDsl7pFQoSZTz5N+2NqRXs8GBwnAwo3ZNxqhQ==}

@@ -6906,7 +6923,7 @@ snapshots:
      '@push.rocks/smartshell': 3.3.0
      tsx: 4.21.0

-  '@git.zone/tstest@3.1.4(@aws-sdk/credential-providers@3.808.0)(socks@2.8.4)(typescript@5.9.3)':
+  '@git.zone/tstest@3.1.6(@aws-sdk/credential-providers@3.808.0)(socks@2.8.4)(typescript@5.9.3)':
    dependencies:
      '@api.global/typedserver': 3.0.80
      '@git.zone/tsbundle': 2.8.1
@@ -7060,7 +7077,7 @@ snapshots:
      '@inquirer/figures': 1.0.15
      '@inquirer/type': 2.0.0
      '@types/mute-stream': 0.0.4
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7
      '@types/wrap-ansi': 3.0.0
      ansi-escapes: 4.3.2
      cli-width: 4.1.0
@@ -7350,6 +7367,11 @@ snapshots:
    dependencies:
      '@lit-labs/ssr-dom-shim': 1.4.0

+  '@mistralai/mistralai@1.12.0':
+    dependencies:
+      zod: 3.25.76
+      zod-to-json-schema: 3.25.1(zod@3.25.76)
+
  '@mixmark-io/domino@2.2.0': {}

  '@module-federation/error-codes@0.22.0': {}
@@ -9635,27 +9657,27 @@ snapshots:

  '@types/bn.js@5.2.0':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/body-parser@1.19.6':
    dependencies:
      '@types/connect': 3.4.38
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/buffer-json@2.0.3': {}

  '@types/clean-css@4.2.11':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7
      source-map: 0.6.1

  '@types/connect@3.4.38':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/cors@2.8.19':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/debug@4.1.12':
    dependencies:
@@ -9665,7 +9687,7 @@ snapshots:

  '@types/dns-packet@5.6.5':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/elliptic@6.4.18':
    dependencies:
@@ -9673,7 +9695,7 @@ snapshots:

  '@types/express-serve-static-core@5.0.7':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7
      '@types/qs': 6.14.0
      '@types/range-parser': 1.2.7
      '@types/send': 0.17.5
@@ -9697,16 +9719,16 @@ snapshots:
  '@types/fs-extra@11.0.4':
    dependencies:
      '@types/jsonfile': 6.1.4
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/fs-extra@9.0.13':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/glob@7.2.0':
    dependencies:
      '@types/minimatch': 5.1.2
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/hast@3.0.4':
    dependencies:
@@ -9728,7 +9750,7 @@ snapshots:

  '@types/jsonfile@6.1.4':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/mdast@4.0.4':
    dependencies:
@@ -9746,18 +9768,22 @@ snapshots:

  '@types/mute-stream@0.0.4':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/node-forge@1.3.11':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/node@16.9.1': {}

-  '@types/node@22.15.17':
+  '@types/node@22.19.7':
    dependencies:
      undici-types: 6.21.0

+  '@types/node@25.0.9':
+    dependencies:
+      undici-types: 7.16.0
+
  '@types/ping@0.4.4': {}

  '@types/qs@6.14.0': {}
@@ -9773,28 +9799,28 @@ snapshots:
  '@types/send@0.17.5':
    dependencies:
      '@types/mime': 1.3.5
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/serve-static@1.15.8':
    dependencies:
      '@types/http-errors': 2.0.5
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7
      '@types/send': 0.17.5

  '@types/serve-static@2.2.0':
    dependencies:
      '@types/http-errors': 2.0.5
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/symbol-tree@3.2.5': {}

  '@types/tar-stream@3.1.4':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/through2@2.0.41':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/trusted-types@2.0.7': {}

@@ -9816,7 +9842,7 @@ snapshots:

  '@types/whatwg-url@8.2.2':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7
      '@types/webidl-conversions': 7.0.3

  '@types/which@3.0.4': {}
@@ -9825,11 +9851,11 @@ snapshots:

  '@types/ws@8.18.1':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7

  '@types/yauzl@2.10.3':
    dependencies:
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7
    optional: true

  '@ungap/structured-clone@1.3.0': {}
@@ -10289,7 +10315,7 @@ snapshots:
  engine.io@6.6.4:
    dependencies:
      '@types/cors': 2.8.19
-      '@types/node': 22.15.17
+      '@types/node': 22.19.7
      accepts: 1.3.8
      base64id: 2.0.0
      cookie: 0.7.2
@@ -11556,7 +11582,7 @@ snapshots:
      is-docker: 2.2.1
      is-wsl: 2.2.0

-  openai@5.12.2(ws@8.18.3)(zod@3.25.76):
+  openai@6.16.0(ws@8.18.3)(zod@3.25.76):
    optionalDependencies:
      ws: 8.18.3
      zod: 3.25.76
@@ -12351,6 +12377,8 @@ snapshots:

  undici-types@6.21.0: {}

+  undici-types@7.16.0: {}
+
  unified@11.0.5:
    dependencies:
      '@types/unist': 3.0.3
@@ -12506,6 +12534,10 @@ snapshots:

  yoctocolors-cjs@2.1.3: {}

+  zod-to-json-schema@3.25.1(zod@3.25.76):
+    dependencies:
+      zod: 3.25.76
+
  zod@3.24.2: {}

  zod@3.25.76: {}
--- a/readme.hints.md
+++ b/readme.hints.md
@@ -3,8 +3,10 @@
 ## Dependencies

 - Uses `@git.zone/tstest` v3.x for testing (import from `@git.zone/tstest/tapbundle`)
- `@push.rocks/smartfile` is kept at v11 to avoid migration to factory pattern
+- `@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
@@ -12,6 +14,68 @@
 - 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
--- a/readme.md
+++ b/readme.md
@@ -6,7 +6,7 @@
 [![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

@@ -58,6 +58,7 @@ Choose the right provider for your use case:
 | -------------- | :--: | :-------: | :-: | :----: | :-------: | :------: | :----: | --------------------------------------------------------------- |
 | **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               |
@@ -282,6 +283,38 @@ const response = await anthropic.chat({
 - 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:
@@ -645,6 +678,7 @@ export ELEVENLABS_API_KEY=sk-...
 | --------------------- | -------------------- | --------------------------------------------------------- |
 | **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) |
--- a/test/test.audio.elevenlabs.ts
+++ b/test/test.audio.elevenlabs.ts
@@ -1,8 +1,9 @@
 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 '@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.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.document.anthropic.ts
+++ b/test/test.document.anthropic.ts
@@ -1,9 +1,10 @@
 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 '@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.vision.anthropic.ts
+++ b/test/test.vision.anthropic.ts
@@ -1,8 +1,9 @@
 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: '0.9.0',
+  version: '0.11.0',
  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/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.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
@@ -12,10 +12,60 @@ import type {
  ImageResponse
 } 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
+}
+
+/**
+ * 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 +73,8 @@ 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 +82,8 @@ 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() {
@@ -154,7 +208,7 @@ export class OllamaProvider extends MultiModalModel {
      { role: 'user', content: optionsArg.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 +217,10 @@ 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 +235,143 @@ export class OllamaProvider extends MultiModalModel {
    };
  }

+  /**
+   * 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 };
+
+    const messages = [
+      { role: 'system', content: optionsArg.systemMessage },
+      ...optionsArg.messageHistory,
+      { role: 'user', content: optionsArg.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 };
+
+    const messages = [
+      { role: 'system', content: optionsArg.systemMessage },
+      ...optionsArg.messageHistory,
+      { role: 'user', content: optionsArg.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.');
  }
Author	SHA1	Message	Date
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