Compare commits
8 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 0403443634 | |||
| e2ed429aac | |||
| 5c856ec3ed | |||
| 052f37294d | |||
| 93bb375059 | |||
| 574f7a594c | |||
| 0b2a058550 | |||
| 88d15c89e5 |
50
changelog.md
50
changelog.md
@@ -1,5 +1,55 @@
|
||||
# Changelog
|
||||
|
||||
## 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
|
||||
- Implement research() for OpenAiProvider (deep research model selection, optional web search/tools, background flag, source extraction)
|
||||
- Implement research() for AnthropicProvider (web search tool support, domain filters, citation extraction)
|
||||
- Implement research() for PerplexityProvider (sonar / sonar-pro model usage and citation parsing)
|
||||
- Add research() stubs to Exo, Groq, Ollama and XAI providers that throw a clear 'not yet supported' error to preserve interface compatibility
|
||||
- Add tests for research interfaces and provider research methods (test files updated/added)
|
||||
- Add documentation: readme.research.md describing the research API, usage and configuration
|
||||
- Export additional providers from ts/index.ts and update provider typings/imports across files
|
||||
- Add a 'typecheck' script to package.json
|
||||
- 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'
|
||||
- Upgraded @anthropic-ai/sdk from ^0.57.0 to ^0.59.0
|
||||
- Upgraded openai from ^5.11.0 to ^5.12.2
|
||||
- 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
|
||||
- Enhanced PDF conversion with improved scale options and quality controls
|
||||
- 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
|
||||
- Fixed memory leaks by properly implementing cleanup in the base class stop() method
|
||||
- Updated SmartAi class to properly stop all providers on shutdown
|
||||
- Updated @push.rocks/smartrequest from v2.1.0 to v4.2.1 with migration to new API
|
||||
- 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
|
||||
|
||||
|
||||
13
package.json
13
package.json
@@ -1,6 +1,6 @@
|
||||
{
|
||||
"name": "@push.rocks/smartai",
|
||||
"version": "0.5.5",
|
||||
"version": "0.6.0",
|
||||
"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",
|
||||
@@ -10,6 +10,7 @@
|
||||
"license": "MIT",
|
||||
"scripts": {
|
||||
"test": "(tstest test/ --web --verbose)",
|
||||
"typecheck": "tsbuild check",
|
||||
"build": "(tsbuild --web --allowimplicitany)",
|
||||
"buildDocs": "(tsdoc)"
|
||||
},
|
||||
@@ -23,15 +24,15 @@
|
||||
"@types/node": "^22.15.17"
|
||||
},
|
||||
"dependencies": {
|
||||
"@anthropic-ai/sdk": "^0.57.0",
|
||||
"@anthropic-ai/sdk": "^0.59.0",
|
||||
"@push.rocks/smartarray": "^1.1.0",
|
||||
"@push.rocks/smartfile": "^11.2.5",
|
||||
"@push.rocks/smartpath": "^5.0.18",
|
||||
"@push.rocks/smartpdf": "^3.2.2",
|
||||
"@push.rocks/smartpath": "^6.0.0",
|
||||
"@push.rocks/smartpdf": "^4.1.1",
|
||||
"@push.rocks/smartpromise": "^4.2.3",
|
||||
"@push.rocks/smartrequest": "^2.1.0",
|
||||
"@push.rocks/smartrequest": "^4.2.1",
|
||||
"@push.rocks/webstream": "^1.0.10",
|
||||
"openai": "^5.10.2"
|
||||
"openai": "^5.12.2"
|
||||
},
|
||||
"repository": {
|
||||
"type": "git",
|
||||
|
||||
1354
pnpm-lock.yaml
generated
1354
pnpm-lock.yaml
generated
File diff suppressed because it is too large
Load Diff
684
readme.md
684
readme.md
@@ -1,393 +1,467 @@
|
||||
# @push.rocks/smartai
|
||||
**One API to rule them all** 🚀
|
||||
|
||||
SmartAi is a powerful TypeScript library that provides a unified interface for integrating with multiple AI providers including OpenAI, Anthropic, Perplexity, Ollama, Groq, XAI, and Exo. It offers comprehensive support for chat interactions, streaming conversations, text-to-speech, document analysis, and vision processing.
|
||||
[](https://www.npmjs.com/package/@push.rocks/smartai)
|
||||
[](https://www.typescriptlang.org/)
|
||||
[](https://opensource.org/licenses/MIT)
|
||||
|
||||
## Install
|
||||
SmartAI unifies the world's leading AI providers - OpenAI, Anthropic, Perplexity, Ollama, Groq, XAI, and Exo - under a single, elegant TypeScript interface. Build AI applications at lightning speed without vendor lock-in.
|
||||
|
||||
To install SmartAi into your project, use pnpm:
|
||||
## 🎯 Why SmartAI?
|
||||
|
||||
- **🔌 Universal Interface**: Write once, run with any AI provider. Switch between GPT-4, Claude, Llama, or Grok with a single line change.
|
||||
- **🛡️ Type-Safe**: Full TypeScript support with comprehensive type definitions for all operations
|
||||
- **🌊 Streaming First**: Built for real-time applications with native streaming support
|
||||
- **🎨 Multi-Modal**: Seamlessly work with text, images, audio, and documents
|
||||
- **🏠 Local & Cloud**: Support for both cloud providers and local models via Ollama
|
||||
- **⚡ Zero Lock-In**: Your code remains portable across all AI providers
|
||||
|
||||
## 🚀 Quick Start
|
||||
|
||||
```bash
|
||||
pnpm install @push.rocks/smartai
|
||||
npm install @push.rocks/smartai
|
||||
```
|
||||
|
||||
## Usage
|
||||
|
||||
SmartAi provides a clean, consistent API across all supported AI providers. This documentation covers all features with practical examples for each provider and capability.
|
||||
|
||||
### Initialization
|
||||
|
||||
First, initialize SmartAi with the API tokens and configuration for the providers you want to use:
|
||||
|
||||
```typescript
|
||||
import { SmartAi } from '@push.rocks/smartai';
|
||||
|
||||
const smartAi = new SmartAi({
|
||||
// OpenAI - for GPT models, DALL-E, and TTS
|
||||
openaiToken: 'your-openai-api-key',
|
||||
|
||||
// Anthropic - for Claude models
|
||||
anthropicToken: 'your-anthropic-api-key',
|
||||
|
||||
// Perplexity - for research-focused AI
|
||||
perplexityToken: 'your-perplexity-api-key',
|
||||
|
||||
// Groq - for fast inference
|
||||
groqToken: 'your-groq-api-key',
|
||||
|
||||
// XAI - for Grok models
|
||||
xaiToken: 'your-xai-api-key',
|
||||
|
||||
// Ollama - for local models
|
||||
ollama: {
|
||||
baseUrl: 'http://localhost:11434',
|
||||
model: 'llama2', // default model for chat
|
||||
visionModel: 'llava' // default model for vision
|
||||
},
|
||||
|
||||
// Exo - for distributed inference
|
||||
exo: {
|
||||
baseUrl: 'http://localhost:8080/v1',
|
||||
apiKey: 'your-exo-api-key'
|
||||
}
|
||||
// Initialize with your favorite providers
|
||||
const ai = new SmartAi({
|
||||
openaiToken: 'sk-...',
|
||||
anthropicToken: 'sk-ant-...'
|
||||
});
|
||||
|
||||
// Start the SmartAi instance
|
||||
await smartAi.start();
|
||||
```
|
||||
await ai.start();
|
||||
|
||||
## Supported Providers
|
||||
|
||||
SmartAi supports the following AI providers:
|
||||
|
||||
| Provider | Use Case | Key Features |
|
||||
|----------|----------|--------------|
|
||||
| **OpenAI** | General purpose, GPT models | Chat, streaming, TTS, vision, documents |
|
||||
| **Anthropic** | Claude models, safety-focused | Chat, streaming, vision, documents |
|
||||
| **Perplexity** | Research and factual queries | Chat, streaming, documents |
|
||||
| **Groq** | Fast inference | Chat, streaming |
|
||||
| **XAI** | Grok models | Chat, streaming |
|
||||
| **Ollama** | Local models | Chat, streaming, vision |
|
||||
| **Exo** | Distributed inference | Chat, streaming |
|
||||
|
||||
## Core Features
|
||||
|
||||
### 1. Chat Interactions
|
||||
|
||||
SmartAi provides both synchronous and streaming chat capabilities across all supported providers.
|
||||
|
||||
#### Synchronous Chat
|
||||
|
||||
Simple request-response interactions with any provider:
|
||||
|
||||
```typescript
|
||||
// OpenAI Example
|
||||
const openAiResponse = await smartAi.openaiProvider.chat({
|
||||
// Same API, multiple providers
|
||||
const response = await ai.openaiProvider.chat({
|
||||
systemMessage: 'You are a helpful assistant.',
|
||||
userMessage: 'What is the capital of France?',
|
||||
userMessage: 'Explain quantum computing in simple terms',
|
||||
messageHistory: []
|
||||
});
|
||||
console.log(openAiResponse.message); // "The capital of France is Paris."
|
||||
|
||||
// Anthropic Example
|
||||
const anthropicResponse = await smartAi.anthropicProvider.chat({
|
||||
systemMessage: 'You are a knowledgeable historian.',
|
||||
userMessage: 'Tell me about the French Revolution',
|
||||
messageHistory: []
|
||||
});
|
||||
console.log(anthropicResponse.message);
|
||||
|
||||
// Using message history for context
|
||||
const contextualResponse = await smartAi.openaiProvider.chat({
|
||||
systemMessage: 'You are a math tutor.',
|
||||
userMessage: 'What about multiplication?',
|
||||
messageHistory: [
|
||||
{ role: 'user', content: 'Can you teach me math?' },
|
||||
{ role: 'assistant', content: 'Of course! What would you like to learn?' }
|
||||
]
|
||||
});
|
||||
```
|
||||
|
||||
#### Streaming Chat
|
||||
## 📊 Provider Capabilities Matrix
|
||||
|
||||
For real-time, token-by-token responses:
|
||||
Choose the right provider for your use case:
|
||||
|
||||
| Provider | Chat | Streaming | TTS | Vision | Documents | Highlights |
|
||||
|----------|:----:|:---------:|:---:|:------:|:---------:|------------|
|
||||
| **OpenAI** | ✅ | ✅ | ✅ | ✅ | ✅ | • GPT-4, DALL-E 3<br>• Industry standard<br>• Most features |
|
||||
| **Anthropic** | ✅ | ✅ | ❌ | ✅ | ✅ | • Claude 3 Opus<br>• Superior reasoning<br>• 200k context |
|
||||
| **Ollama** | ✅ | ✅ | ❌ | ✅ | ✅ | • 100% local<br>• Privacy-first<br>• No API costs |
|
||||
| **XAI** | ✅ | ✅ | ❌ | ❌ | ✅ | • Grok models<br>• Real-time data<br>• Uncensored |
|
||||
| **Perplexity** | ✅ | ✅ | ❌ | ❌ | ❌ | • Web-aware<br>• Research-focused<br>• Citations |
|
||||
| **Groq** | ✅ | ✅ | ❌ | ❌ | ❌ | • 10x faster<br>• LPU inference<br>• Low latency |
|
||||
| **Exo** | ✅ | ✅ | ❌ | ❌ | ❌ | • Distributed<br>• P2P compute<br>• Decentralized |
|
||||
|
||||
## 🎮 Core Features
|
||||
|
||||
### 💬 Universal Chat Interface
|
||||
|
||||
Works identically across all providers:
|
||||
|
||||
```typescript
|
||||
// Create a readable stream for input
|
||||
const { readable, writable } = new TransformStream();
|
||||
const writer = writable.getWriter();
|
||||
// Use GPT-4 for complex reasoning
|
||||
const gptResponse = await ai.openaiProvider.chat({
|
||||
systemMessage: 'You are a expert physicist.',
|
||||
userMessage: 'Explain the implications of quantum entanglement',
|
||||
messageHistory: []
|
||||
});
|
||||
|
||||
// Send a message
|
||||
const encoder = new TextEncoder();
|
||||
await writer.write(encoder.encode(JSON.stringify({
|
||||
role: 'user',
|
||||
content: 'Write a haiku about programming'
|
||||
})));
|
||||
await writer.close();
|
||||
// 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: []
|
||||
});
|
||||
|
||||
// Get streaming response
|
||||
const responseStream = await smartAi.openaiProvider.chatStream(readable);
|
||||
const reader = responseStream.getReader();
|
||||
const decoder = new TextDecoder();
|
||||
// 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: []
|
||||
});
|
||||
```
|
||||
|
||||
// Read the stream
|
||||
### 🌊 Real-Time Streaming
|
||||
|
||||
Build responsive chat interfaces with token-by-token streaming:
|
||||
|
||||
```typescript
|
||||
// Create a chat stream
|
||||
const stream = await ai.openaiProvider.chatStream(inputStream);
|
||||
const reader = stream.getReader();
|
||||
|
||||
// Display responses as they arrive
|
||||
while (true) {
|
||||
const { done, value } = await reader.read();
|
||||
if (done) break;
|
||||
process.stdout.write(value); // Print each chunk as it arrives
|
||||
|
||||
// Update UI in real-time
|
||||
process.stdout.write(value);
|
||||
}
|
||||
```
|
||||
|
||||
### 2. Text-to-Speech (Audio Generation)
|
||||
### 🎙️ Text-to-Speech
|
||||
|
||||
Convert text to natural-sounding speech (currently supported by OpenAI):
|
||||
Generate natural voices with OpenAI:
|
||||
|
||||
```typescript
|
||||
import * as fs from 'fs';
|
||||
|
||||
// Generate speech from text
|
||||
const audioStream = await smartAi.openaiProvider.audio({
|
||||
message: 'Hello world! This is a test of the text-to-speech system.'
|
||||
const audioStream = await ai.openaiProvider.audio({
|
||||
message: 'Welcome to the future of AI development!'
|
||||
});
|
||||
|
||||
// Save to file
|
||||
const writeStream = fs.createWriteStream('output.mp3');
|
||||
audioStream.pipe(writeStream);
|
||||
// Stream directly to speakers
|
||||
audioStream.pipe(speakerOutput);
|
||||
|
||||
// Or use in your application directly
|
||||
audioStream.on('data', (chunk) => {
|
||||
// Process audio chunks
|
||||
// Or save to file
|
||||
audioStream.pipe(fs.createWriteStream('welcome.mp3'));
|
||||
```
|
||||
|
||||
### 👁️ Vision Analysis
|
||||
|
||||
Understand images with multiple providers:
|
||||
|
||||
```typescript
|
||||
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'
|
||||
});
|
||||
|
||||
// Anthropic: Detailed analysis
|
||||
const claudeVision = await ai.anthropicProvider.vision({
|
||||
image,
|
||||
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'
|
||||
});
|
||||
```
|
||||
|
||||
### 3. Vision Processing
|
||||
### 📄 Document Intelligence
|
||||
|
||||
Analyze images and get detailed descriptions:
|
||||
Extract insights from PDFs with AI:
|
||||
|
||||
```typescript
|
||||
import * as fs from 'fs';
|
||||
const contract = fs.readFileSync('contract.pdf');
|
||||
const invoice = fs.readFileSync('invoice.pdf');
|
||||
|
||||
// Read an image file
|
||||
const imageBuffer = fs.readFileSync('image.jpg');
|
||||
|
||||
// OpenAI Vision
|
||||
const openAiVision = await smartAi.openaiProvider.vision({
|
||||
image: imageBuffer,
|
||||
prompt: 'What is in this image? Describe in detail.'
|
||||
});
|
||||
console.log('OpenAI:', openAiVision);
|
||||
|
||||
// Anthropic Vision
|
||||
const anthropicVision = await smartAi.anthropicProvider.vision({
|
||||
image: imageBuffer,
|
||||
prompt: 'Analyze this image and identify any text or objects.'
|
||||
});
|
||||
console.log('Anthropic:', anthropicVision);
|
||||
|
||||
// Ollama Vision (using local model)
|
||||
const ollamaVision = await smartAi.ollamaProvider.vision({
|
||||
image: imageBuffer,
|
||||
prompt: 'Describe the colors and composition of this image.'
|
||||
});
|
||||
console.log('Ollama:', ollamaVision);
|
||||
```
|
||||
|
||||
### 4. Document Analysis
|
||||
|
||||
Process and analyze PDF documents with AI:
|
||||
|
||||
```typescript
|
||||
import * as fs from 'fs';
|
||||
|
||||
// Read PDF documents
|
||||
const pdfBuffer = fs.readFileSync('document.pdf');
|
||||
|
||||
// Analyze with OpenAI
|
||||
const openAiAnalysis = await smartAi.openaiProvider.document({
|
||||
systemMessage: 'You are a document analyst. Extract key information.',
|
||||
userMessage: 'Summarize this document and list the main points.',
|
||||
messageHistory: [],
|
||||
pdfDocuments: [pdfBuffer]
|
||||
});
|
||||
console.log('OpenAI Analysis:', openAiAnalysis.message);
|
||||
|
||||
// Analyze with Anthropic
|
||||
const anthropicAnalysis = await smartAi.anthropicProvider.document({
|
||||
// Analyze documents
|
||||
const analysis = await ai.openaiProvider.document({
|
||||
systemMessage: 'You are a legal expert.',
|
||||
userMessage: 'Identify any legal terms or implications in this document.',
|
||||
userMessage: 'Compare these documents and highlight key differences',
|
||||
messageHistory: [],
|
||||
pdfDocuments: [pdfBuffer]
|
||||
pdfDocuments: [contract, invoice]
|
||||
});
|
||||
console.log('Anthropic Analysis:', anthropicAnalysis.message);
|
||||
|
||||
// Process multiple documents
|
||||
const doc1 = fs.readFileSync('contract1.pdf');
|
||||
const doc2 = fs.readFileSync('contract2.pdf');
|
||||
|
||||
const comparison = await smartAi.openaiProvider.document({
|
||||
systemMessage: 'You are a contract analyst.',
|
||||
userMessage: 'Compare these two contracts and highlight the differences.',
|
||||
// Multi-document analysis
|
||||
const taxDocs = [form1099, w2, receipts];
|
||||
const taxAnalysis = await ai.anthropicProvider.document({
|
||||
systemMessage: 'You are a tax advisor.',
|
||||
userMessage: 'Prepare a tax summary from these documents',
|
||||
messageHistory: [],
|
||||
pdfDocuments: [doc1, doc2]
|
||||
pdfDocuments: taxDocs
|
||||
});
|
||||
console.log('Comparison:', comparison.message);
|
||||
```
|
||||
|
||||
### 5. Conversation Management
|
||||
### 🔄 Persistent Conversations
|
||||
|
||||
Create persistent conversation sessions with any provider:
|
||||
Maintain context across interactions:
|
||||
|
||||
```typescript
|
||||
// Create a conversation with OpenAI
|
||||
const conversation = smartAi.createConversation('openai');
|
||||
// Create a coding assistant conversation
|
||||
const assistant = ai.createConversation('openai');
|
||||
await assistant.setSystemMessage('You are an expert TypeScript developer.');
|
||||
|
||||
// Set the system message
|
||||
await conversation.setSystemMessage('You are a helpful coding assistant.');
|
||||
|
||||
// Get input and output streams
|
||||
const inputWriter = conversation.getInputStreamWriter();
|
||||
const outputStream = conversation.getOutputStream();
|
||||
|
||||
// Set up output reader
|
||||
const reader = outputStream.getReader();
|
||||
const decoder = new TextDecoder();
|
||||
|
||||
// Send messages
|
||||
await inputWriter.write('How do I create a REST API in Node.js?');
|
||||
|
||||
// Read responses
|
||||
while (true) {
|
||||
const { done, value } = await reader.read();
|
||||
if (done) break;
|
||||
console.log('Assistant:', decoder.decode(value));
|
||||
}
|
||||
// First question
|
||||
const inputWriter = assistant.getInputStreamWriter();
|
||||
await inputWriter.write('How do I implement a singleton pattern?');
|
||||
|
||||
// Continue the conversation
|
||||
await inputWriter.write('Can you show me an example with Express?');
|
||||
await inputWriter.write('Now show me how to make it thread-safe');
|
||||
|
||||
// Create conversations with different providers
|
||||
const anthropicConversation = smartAi.createConversation('anthropic');
|
||||
const groqConversation = smartAi.createConversation('groq');
|
||||
// The assistant remembers the entire context
|
||||
```
|
||||
|
||||
## Advanced Usage
|
||||
## 🚀 Real-World Examples
|
||||
|
||||
### Error Handling
|
||||
|
||||
Always wrap AI operations in try-catch blocks for robust error handling:
|
||||
### Build a Customer Support Bot
|
||||
|
||||
```typescript
|
||||
try {
|
||||
const response = await smartAi.openaiProvider.chat({
|
||||
systemMessage: 'You are an assistant.',
|
||||
userMessage: 'Hello!',
|
||||
const supportBot = new SmartAi({
|
||||
anthropicToken: process.env.ANTHROPIC_KEY // Claude for empathetic responses
|
||||
});
|
||||
|
||||
async function handleCustomerQuery(query: string, history: ChatMessage[]) {
|
||||
try {
|
||||
const response = await supportBot.anthropicProvider.chat({
|
||||
systemMessage: `You are a helpful customer support agent.
|
||||
Be empathetic, professional, and solution-oriented.`,
|
||||
userMessage: query,
|
||||
messageHistory: history
|
||||
});
|
||||
|
||||
return response.message;
|
||||
} catch (error) {
|
||||
// Fallback to another provider if needed
|
||||
return await supportBot.openaiProvider.chat({...});
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
### Create a Code Review Assistant
|
||||
|
||||
```typescript
|
||||
const codeReviewer = new SmartAi({
|
||||
groqToken: process.env.GROQ_KEY // Groq for speed
|
||||
});
|
||||
|
||||
async function reviewCode(code: string, language: string) {
|
||||
const startTime = Date.now();
|
||||
|
||||
const review = await codeReviewer.groqProvider.chat({
|
||||
systemMessage: `You are a ${language} expert. Review code for:
|
||||
- Security vulnerabilities
|
||||
- Performance issues
|
||||
- Best practices
|
||||
- Potential bugs`,
|
||||
userMessage: `Review this code:\n\n${code}`,
|
||||
messageHistory: []
|
||||
});
|
||||
console.log(response.message);
|
||||
} catch (error) {
|
||||
if (error.code === 'rate_limit_exceeded') {
|
||||
console.error('Rate limit hit, please retry later');
|
||||
} else if (error.code === 'invalid_api_key') {
|
||||
console.error('Invalid API key provided');
|
||||
} else {
|
||||
console.error('Unexpected error:', error.message);
|
||||
|
||||
console.log(`Review completed in ${Date.now() - startTime}ms`);
|
||||
return review.message;
|
||||
}
|
||||
```
|
||||
|
||||
### Build a Research Assistant
|
||||
|
||||
```typescript
|
||||
const researcher = new SmartAi({
|
||||
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.',
|
||||
userMessage: `Research the latest developments in ${topic}`,
|
||||
messageHistory: []
|
||||
});
|
||||
|
||||
return findings.message;
|
||||
}
|
||||
```
|
||||
|
||||
### Local AI for Sensitive Data
|
||||
|
||||
```typescript
|
||||
const localAI = new SmartAi({
|
||||
ollama: {
|
||||
baseUrl: 'http://localhost:11434',
|
||||
model: 'llama2',
|
||||
visionModel: 'llava'
|
||||
}
|
||||
});
|
||||
|
||||
// Process sensitive documents without leaving your infrastructure
|
||||
async function analyzeSensitiveDoc(pdfBuffer: Buffer) {
|
||||
const analysis = await localAI.ollamaProvider.document({
|
||||
systemMessage: 'Extract and summarize key information.',
|
||||
userMessage: 'Analyze this confidential document',
|
||||
messageHistory: [],
|
||||
pdfDocuments: [pdfBuffer]
|
||||
});
|
||||
|
||||
// Data never leaves your servers
|
||||
return analysis.message;
|
||||
}
|
||||
```
|
||||
|
||||
## ⚡ Performance Tips
|
||||
|
||||
### 1. Provider Selection Strategy
|
||||
|
||||
```typescript
|
||||
class SmartAIRouter {
|
||||
constructor(private ai: SmartAi) {}
|
||||
|
||||
async query(message: string, requirements: {
|
||||
speed?: boolean;
|
||||
accuracy?: boolean;
|
||||
cost?: boolean;
|
||||
privacy?: boolean;
|
||||
}) {
|
||||
if (requirements.privacy) {
|
||||
return this.ai.ollamaProvider.chat({...}); // Local only
|
||||
}
|
||||
if (requirements.speed) {
|
||||
return this.ai.groqProvider.chat({...}); // 10x faster
|
||||
}
|
||||
if (requirements.accuracy) {
|
||||
return this.ai.anthropicProvider.chat({...}); // Best reasoning
|
||||
}
|
||||
// Default fallback
|
||||
return this.ai.openaiProvider.chat({...});
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
### Streaming with Custom Processing
|
||||
|
||||
Implement custom transformations on streaming responses:
|
||||
### 2. Streaming for Large Responses
|
||||
|
||||
```typescript
|
||||
// Create a custom transform stream
|
||||
const customTransform = new TransformStream({
|
||||
transform(chunk, controller) {
|
||||
// Example: Add timestamps to each chunk
|
||||
const timestamp = new Date().toISOString();
|
||||
controller.enqueue(`[${timestamp}] ${chunk}`);
|
||||
// Don't wait for the entire response
|
||||
async function streamResponse(userQuery: string) {
|
||||
const stream = await ai.openaiProvider.chatStream(createInputStream(userQuery));
|
||||
|
||||
// Process tokens as they arrive
|
||||
for await (const chunk of stream) {
|
||||
updateUI(chunk); // Immediate feedback
|
||||
await processChunk(chunk); // Parallel processing
|
||||
}
|
||||
});
|
||||
|
||||
// Apply to streaming chat
|
||||
const inputStream = new ReadableStream({
|
||||
start(controller) {
|
||||
controller.enqueue(new TextEncoder().encode(JSON.stringify({
|
||||
role: 'user',
|
||||
content: 'Tell me a story'
|
||||
})));
|
||||
controller.close();
|
||||
}
|
||||
});
|
||||
|
||||
const responseStream = await smartAi.openaiProvider.chatStream(inputStream);
|
||||
const processedStream = responseStream.pipeThrough(customTransform);
|
||||
|
||||
// Read processed stream
|
||||
const reader = processedStream.getReader();
|
||||
while (true) {
|
||||
const { done, value } = await reader.read();
|
||||
if (done) break;
|
||||
console.log(value);
|
||||
}
|
||||
```
|
||||
|
||||
### Provider-Specific Features
|
||||
|
||||
Each provider may have unique capabilities. Here's how to leverage them:
|
||||
### 3. Parallel Multi-Provider Queries
|
||||
|
||||
```typescript
|
||||
// OpenAI - Use specific models
|
||||
const gpt4Response = await smartAi.openaiProvider.chat({
|
||||
systemMessage: 'You are a helpful assistant.',
|
||||
userMessage: 'Explain quantum computing',
|
||||
messageHistory: []
|
||||
});
|
||||
|
||||
// Anthropic - Use Claude's strength in analysis
|
||||
const codeReview = await smartAi.anthropicProvider.chat({
|
||||
systemMessage: 'You are a code reviewer.',
|
||||
userMessage: 'Review this code for security issues: ...',
|
||||
messageHistory: []
|
||||
});
|
||||
|
||||
// Perplexity - Best for research and current events
|
||||
const research = await smartAi.perplexityProvider.chat({
|
||||
systemMessage: 'You are a research assistant.',
|
||||
userMessage: 'What are the latest developments in renewable energy?',
|
||||
messageHistory: []
|
||||
});
|
||||
|
||||
// Groq - Optimized for speed
|
||||
const quickResponse = await smartAi.groqProvider.chat({
|
||||
systemMessage: 'You are a quick helper.',
|
||||
userMessage: 'Give me a one-line summary of photosynthesis',
|
||||
messageHistory: []
|
||||
});
|
||||
// Get the best answer from multiple AIs
|
||||
async function consensusQuery(question: string) {
|
||||
const providers = [
|
||||
ai.openaiProvider.chat({...}),
|
||||
ai.anthropicProvider.chat({...}),
|
||||
ai.perplexityProvider.chat({...})
|
||||
];
|
||||
|
||||
const responses = await Promise.all(providers);
|
||||
return synthesizeResponses(responses);
|
||||
}
|
||||
```
|
||||
|
||||
### Performance Optimization
|
||||
## 🛠️ Advanced Features
|
||||
|
||||
Tips for optimal performance:
|
||||
### Custom Streaming Transformations
|
||||
|
||||
```typescript
|
||||
// 1. Reuse providers instead of creating new instances
|
||||
const smartAi = new SmartAi({ /* config */ });
|
||||
await smartAi.start(); // Initialize once
|
||||
// Add real-time translation
|
||||
const translationStream = new TransformStream({
|
||||
async transform(chunk, controller) {
|
||||
const translated = await translateChunk(chunk);
|
||||
controller.enqueue(translated);
|
||||
}
|
||||
});
|
||||
|
||||
// 2. Use streaming for long responses
|
||||
// Streaming reduces time-to-first-token and memory usage
|
||||
|
||||
// 3. Batch operations when possible
|
||||
const promises = [
|
||||
smartAi.openaiProvider.chat({ /* ... */ }),
|
||||
smartAi.anthropicProvider.chat({ /* ... */ })
|
||||
];
|
||||
const results = await Promise.all(promises);
|
||||
|
||||
// 4. Clean up resources
|
||||
await smartAi.stop(); // When done
|
||||
const responseStream = await ai.openaiProvider.chatStream(input);
|
||||
const translatedStream = responseStream.pipeThrough(translationStream);
|
||||
```
|
||||
|
||||
### Error Handling & Fallbacks
|
||||
|
||||
```typescript
|
||||
class ResilientAI {
|
||||
private providers = ['openai', 'anthropic', 'groq'];
|
||||
|
||||
async query(opts: ChatOptions): Promise<ChatResponse> {
|
||||
for (const provider of this.providers) {
|
||||
try {
|
||||
return await this.ai[`${provider}Provider`].chat(opts);
|
||||
} catch (error) {
|
||||
console.warn(`${provider} failed, trying next...`);
|
||||
continue;
|
||||
}
|
||||
}
|
||||
throw new Error('All providers failed');
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
### Token Counting & Cost Management
|
||||
|
||||
```typescript
|
||||
// Track usage across providers
|
||||
class UsageTracker {
|
||||
async trackedChat(provider: string, options: ChatOptions) {
|
||||
const start = Date.now();
|
||||
const response = await ai[`${provider}Provider`].chat(options);
|
||||
|
||||
const usage = {
|
||||
provider,
|
||||
duration: Date.now() - start,
|
||||
inputTokens: estimateTokens(options),
|
||||
outputTokens: estimateTokens(response.message)
|
||||
};
|
||||
|
||||
await this.logUsage(usage);
|
||||
return response;
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
## 📦 Installation & Setup
|
||||
|
||||
### Prerequisites
|
||||
|
||||
- Node.js 16+
|
||||
- TypeScript 4.5+
|
||||
- API keys for your chosen providers
|
||||
|
||||
### Environment Setup
|
||||
|
||||
```bash
|
||||
# Install
|
||||
npm install @push.rocks/smartai
|
||||
|
||||
# Set up environment variables
|
||||
export OPENAI_API_KEY=sk-...
|
||||
export ANTHROPIC_API_KEY=sk-ant-...
|
||||
export PERPLEXITY_API_KEY=pplx-...
|
||||
# ... etc
|
||||
```
|
||||
|
||||
### TypeScript Configuration
|
||||
|
||||
```json
|
||||
{
|
||||
"compilerOptions": {
|
||||
"target": "ES2022",
|
||||
"module": "NodeNext",
|
||||
"lib": ["ES2022"],
|
||||
"strict": true,
|
||||
"esModuleInterop": true,
|
||||
"skipLibCheck": true
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
## 🎯 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 |
|
||||
| **Research & Facts** | Perplexity | Web-aware, provides citations |
|
||||
| **Speed Critical** | Groq | 10x faster inference, sub-second responses |
|
||||
| **Privacy Critical** | Ollama | 100% local, no data leaves your servers |
|
||||
| **Real-time Data** | XAI | Access to current information |
|
||||
| **Cost Sensitive** | Ollama/Exo | Free (local) or distributed compute |
|
||||
|
||||
## 📈 Roadmap
|
||||
|
||||
- [ ] Streaming function calls
|
||||
- [ ] Image generation support
|
||||
- [ ] Voice input processing
|
||||
- [ ] Fine-tuning integration
|
||||
- [ ] Embedding support
|
||||
- [ ] Agent framework
|
||||
- [ ] More providers (Cohere, AI21, etc.)
|
||||
|
||||
## 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.
|
||||
@@ -405,4 +479,4 @@ 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.
|
||||
|
||||
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.
|
||||
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.
|
||||
177
readme.research.md
Normal file
177
readme.research.md
Normal file
@@ -0,0 +1,177 @@
|
||||
# SmartAI Research API Implementation
|
||||
|
||||
This document describes the new research capabilities added to the SmartAI library, enabling web search and deep research features for OpenAI and Anthropic providers.
|
||||
|
||||
## Features Added
|
||||
|
||||
### 1. Research Method Interface
|
||||
|
||||
Added a new `research()` method to the `MultiModalModel` abstract class with the following interfaces:
|
||||
|
||||
```typescript
|
||||
interface ResearchOptions {
|
||||
query: string;
|
||||
searchDepth?: 'basic' | 'advanced' | 'deep';
|
||||
maxSources?: number;
|
||||
includeWebSearch?: boolean;
|
||||
background?: boolean;
|
||||
}
|
||||
|
||||
interface ResearchResponse {
|
||||
answer: string;
|
||||
sources: Array<{
|
||||
url: string;
|
||||
title: string;
|
||||
snippet: string;
|
||||
}>;
|
||||
searchQueries?: string[];
|
||||
metadata?: any;
|
||||
}
|
||||
```
|
||||
|
||||
### 2. OpenAI Provider Research Implementation
|
||||
|
||||
The OpenAI provider now supports:
|
||||
- **Deep Research API** with models:
|
||||
- `o3-deep-research-2025-06-26` (comprehensive analysis)
|
||||
- `o4-mini-deep-research-2025-06-26` (lightweight, faster)
|
||||
- **Web Search** for standard models (gpt-5, o3, o3-pro, o4-mini)
|
||||
- **Background processing** for async deep research tasks
|
||||
|
||||
### 3. Anthropic Provider Research Implementation
|
||||
|
||||
The Anthropic provider now supports:
|
||||
- **Web Search API** with Claude models
|
||||
- **Domain filtering** (allow/block lists)
|
||||
- **Progressive searches** for comprehensive research
|
||||
- **Citation extraction** from responses
|
||||
|
||||
### 4. Perplexity Provider Research Implementation
|
||||
|
||||
The Perplexity provider implements research using:
|
||||
- **Sonar models** for standard searches
|
||||
- **Sonar Pro** for deep research
|
||||
- Built-in citation support
|
||||
|
||||
### 5. Other Providers
|
||||
|
||||
Added research method stubs to:
|
||||
- Groq Provider
|
||||
- Ollama Provider
|
||||
- xAI Provider
|
||||
- Exo Provider
|
||||
|
||||
These providers throw a "not yet supported" error when research is called, maintaining interface compatibility.
|
||||
|
||||
## Usage Examples
|
||||
|
||||
### Basic Research with OpenAI
|
||||
|
||||
```typescript
|
||||
import { OpenAiProvider } from '@push.rocks/smartai';
|
||||
|
||||
const openai = new OpenAiProvider({
|
||||
openaiToken: 'your-api-key',
|
||||
researchModel: 'o4-mini-deep-research-2025-06-26'
|
||||
});
|
||||
|
||||
await openai.start();
|
||||
|
||||
const result = await openai.research({
|
||||
query: 'What are the latest developments in quantum computing?',
|
||||
searchDepth: 'basic',
|
||||
includeWebSearch: true
|
||||
});
|
||||
|
||||
console.log(result.answer);
|
||||
console.log('Sources:', result.sources);
|
||||
```
|
||||
|
||||
### Deep Research with OpenAI
|
||||
|
||||
```typescript
|
||||
const deepResult = await openai.research({
|
||||
query: 'Comprehensive analysis of climate change mitigation strategies',
|
||||
searchDepth: 'deep',
|
||||
background: true
|
||||
});
|
||||
```
|
||||
|
||||
### Research with Anthropic
|
||||
|
||||
```typescript
|
||||
import { AnthropicProvider } from '@push.rocks/smartai';
|
||||
|
||||
const anthropic = new AnthropicProvider({
|
||||
anthropicToken: 'your-api-key',
|
||||
enableWebSearch: true,
|
||||
searchDomainAllowList: ['nature.com', 'science.org']
|
||||
});
|
||||
|
||||
await anthropic.start();
|
||||
|
||||
const result = await anthropic.research({
|
||||
query: 'Latest breakthroughs in CRISPR gene editing',
|
||||
searchDepth: 'advanced'
|
||||
});
|
||||
```
|
||||
|
||||
### Research with Perplexity
|
||||
|
||||
```typescript
|
||||
import { PerplexityProvider } from '@push.rocks/smartai';
|
||||
|
||||
const perplexity = new PerplexityProvider({
|
||||
perplexityToken: 'your-api-key'
|
||||
});
|
||||
|
||||
const result = await perplexity.research({
|
||||
query: 'Current state of autonomous vehicle technology',
|
||||
searchDepth: 'deep' // Uses Sonar Pro model
|
||||
});
|
||||
```
|
||||
|
||||
## Configuration Options
|
||||
|
||||
### OpenAI Provider
|
||||
- `researchModel`: Specify deep research model (default: `o4-mini-deep-research-2025-06-26`)
|
||||
- `enableWebSearch`: Enable web search for standard models
|
||||
|
||||
### Anthropic Provider
|
||||
- `enableWebSearch`: Enable web search capabilities
|
||||
- `searchDomainAllowList`: Array of allowed domains
|
||||
- `searchDomainBlockList`: Array of blocked domains
|
||||
|
||||
## API Pricing
|
||||
|
||||
- **OpenAI Deep Research**: $10 per 1,000 calls
|
||||
- **Anthropic Web Search**: $10 per 1,000 searches + standard token costs
|
||||
- **Perplexity Sonar**: $5 per 1,000 searches (Sonar Pro)
|
||||
|
||||
## Testing
|
||||
|
||||
Run the test suite:
|
||||
|
||||
```bash
|
||||
pnpm test test/test.research.ts
|
||||
```
|
||||
|
||||
All providers have been tested to ensure:
|
||||
- Research methods are properly exposed
|
||||
- Interfaces are correctly typed
|
||||
- Unsupported providers throw appropriate errors
|
||||
|
||||
## Next Steps
|
||||
|
||||
Future enhancements could include:
|
||||
1. Implementing Google Gemini Grounding API support
|
||||
2. Adding Brave Search API integration
|
||||
3. Implementing retry logic for rate limits
|
||||
4. Adding caching for repeated queries
|
||||
5. Supporting batch research operations
|
||||
|
||||
## Notes
|
||||
|
||||
- The implementation maintains backward compatibility
|
||||
- All existing methods continue to work unchanged
|
||||
- Research capabilities are optional and don't affect existing functionality
|
||||
92
test/test.basic.ts
Normal file
92
test/test.basic.ts
Normal file
@@ -0,0 +1,92 @@
|
||||
import { tap, expect } from '@push.rocks/tapbundle';
|
||||
import * as smartai from '../ts/index.js';
|
||||
|
||||
// Basic instantiation tests that don't require API tokens
|
||||
// These tests can run in CI/CD environments without credentials
|
||||
|
||||
tap.test('Basic: should create SmartAi instance', async () => {
|
||||
const testSmartai = new smartai.SmartAi({
|
||||
openaiToken: 'dummy-token-for-testing'
|
||||
});
|
||||
expect(testSmartai).toBeInstanceOf(smartai.SmartAi);
|
||||
expect(testSmartai.openaiProvider).toBeTruthy();
|
||||
});
|
||||
|
||||
tap.test('Basic: should instantiate OpenAI provider', async () => {
|
||||
const openaiProvider = new smartai.OpenAiProvider({
|
||||
openaiToken: 'dummy-token'
|
||||
});
|
||||
expect(openaiProvider).toBeInstanceOf(smartai.OpenAiProvider);
|
||||
expect(typeof openaiProvider.chat).toEqual('function');
|
||||
expect(typeof openaiProvider.audio).toEqual('function');
|
||||
expect(typeof openaiProvider.vision).toEqual('function');
|
||||
expect(typeof openaiProvider.document).toEqual('function');
|
||||
expect(typeof openaiProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Basic: should instantiate Anthropic provider', async () => {
|
||||
const anthropicProvider = new smartai.AnthropicProvider({
|
||||
anthropicToken: 'dummy-token'
|
||||
});
|
||||
expect(anthropicProvider).toBeInstanceOf(smartai.AnthropicProvider);
|
||||
expect(typeof anthropicProvider.chat).toEqual('function');
|
||||
expect(typeof anthropicProvider.audio).toEqual('function');
|
||||
expect(typeof anthropicProvider.vision).toEqual('function');
|
||||
expect(typeof anthropicProvider.document).toEqual('function');
|
||||
expect(typeof anthropicProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Basic: should instantiate Perplexity provider', async () => {
|
||||
const perplexityProvider = new smartai.PerplexityProvider({
|
||||
perplexityToken: 'dummy-token'
|
||||
});
|
||||
expect(perplexityProvider).toBeInstanceOf(smartai.PerplexityProvider);
|
||||
expect(typeof perplexityProvider.chat).toEqual('function');
|
||||
expect(typeof perplexityProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Basic: should instantiate Groq provider', async () => {
|
||||
const groqProvider = new smartai.GroqProvider({
|
||||
groqToken: 'dummy-token'
|
||||
});
|
||||
expect(groqProvider).toBeInstanceOf(smartai.GroqProvider);
|
||||
expect(typeof groqProvider.chat).toEqual('function');
|
||||
expect(typeof groqProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Basic: should instantiate Ollama provider', async () => {
|
||||
const ollamaProvider = new smartai.OllamaProvider({
|
||||
baseUrl: 'http://localhost:11434'
|
||||
});
|
||||
expect(ollamaProvider).toBeInstanceOf(smartai.OllamaProvider);
|
||||
expect(typeof ollamaProvider.chat).toEqual('function');
|
||||
expect(typeof ollamaProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Basic: should instantiate xAI provider', async () => {
|
||||
const xaiProvider = new smartai.XaiProvider({
|
||||
xaiToken: 'dummy-token'
|
||||
});
|
||||
expect(xaiProvider).toBeInstanceOf(smartai.XaiProvider);
|
||||
expect(typeof xaiProvider.chat).toEqual('function');
|
||||
expect(typeof xaiProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Basic: should instantiate Exo provider', async () => {
|
||||
const exoProvider = new smartai.ExoProvider({
|
||||
exoBaseUrl: 'http://localhost:8000'
|
||||
});
|
||||
expect(exoProvider).toBeInstanceOf(smartai.ExoProvider);
|
||||
expect(typeof exoProvider.chat).toEqual('function');
|
||||
expect(typeof exoProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Basic: all providers should extend MultiModalModel', async () => {
|
||||
const openai = new smartai.OpenAiProvider({ openaiToken: 'test' });
|
||||
const anthropic = new smartai.AnthropicProvider({ anthropicToken: 'test' });
|
||||
|
||||
expect(openai).toBeInstanceOf(smartai.MultiModalModel);
|
||||
expect(anthropic).toBeInstanceOf(smartai.MultiModalModel);
|
||||
});
|
||||
|
||||
export default tap.start();
|
||||
140
test/test.interfaces.ts
Normal file
140
test/test.interfaces.ts
Normal file
@@ -0,0 +1,140 @@
|
||||
import { tap, expect } from '@push.rocks/tapbundle';
|
||||
import * as smartai from '../ts/index.js';
|
||||
|
||||
// Test interface exports and type checking
|
||||
// These tests verify that all interfaces are properly exported and usable
|
||||
|
||||
tap.test('Interfaces: ResearchOptions should be properly typed', async () => {
|
||||
const testOptions: smartai.ResearchOptions = {
|
||||
query: 'test query',
|
||||
searchDepth: 'basic',
|
||||
maxSources: 10,
|
||||
includeWebSearch: true,
|
||||
background: false
|
||||
};
|
||||
|
||||
expect(testOptions).toBeInstanceOf(Object);
|
||||
expect(testOptions.query).toEqual('test query');
|
||||
expect(testOptions.searchDepth).toEqual('basic');
|
||||
});
|
||||
|
||||
tap.test('Interfaces: ResearchResponse should be properly typed', async () => {
|
||||
const testResponse: smartai.ResearchResponse = {
|
||||
answer: 'test answer',
|
||||
sources: [
|
||||
{
|
||||
url: 'https://example.com',
|
||||
title: 'Example Source',
|
||||
snippet: 'This is a snippet'
|
||||
}
|
||||
],
|
||||
searchQueries: ['query1', 'query2'],
|
||||
metadata: {
|
||||
model: 'test-model',
|
||||
tokensUsed: 100
|
||||
}
|
||||
};
|
||||
|
||||
expect(testResponse).toBeInstanceOf(Object);
|
||||
expect(testResponse.answer).toEqual('test answer');
|
||||
expect(testResponse.sources).toBeArray();
|
||||
expect(testResponse.sources[0].url).toEqual('https://example.com');
|
||||
});
|
||||
|
||||
tap.test('Interfaces: ChatOptions should be properly typed', async () => {
|
||||
const testChatOptions: smartai.ChatOptions = {
|
||||
systemMessage: 'You are a helpful assistant',
|
||||
userMessage: 'Hello',
|
||||
messageHistory: [
|
||||
{ role: 'user', content: 'Previous message' },
|
||||
{ role: 'assistant', content: 'Previous response' }
|
||||
]
|
||||
};
|
||||
|
||||
expect(testChatOptions).toBeInstanceOf(Object);
|
||||
expect(testChatOptions.systemMessage).toBeTruthy();
|
||||
expect(testChatOptions.messageHistory).toBeArray();
|
||||
});
|
||||
|
||||
tap.test('Interfaces: ChatResponse should be properly typed', async () => {
|
||||
const testChatResponse: smartai.ChatResponse = {
|
||||
role: 'assistant',
|
||||
message: 'This is a response'
|
||||
};
|
||||
|
||||
expect(testChatResponse).toBeInstanceOf(Object);
|
||||
expect(testChatResponse.role).toEqual('assistant');
|
||||
expect(testChatResponse.message).toBeTruthy();
|
||||
});
|
||||
|
||||
tap.test('Interfaces: ChatMessage should be properly typed', async () => {
|
||||
const testMessage: smartai.ChatMessage = {
|
||||
role: 'user',
|
||||
content: 'Test message'
|
||||
};
|
||||
|
||||
expect(testMessage).toBeInstanceOf(Object);
|
||||
expect(testMessage.role).toBeOneOf(['user', 'assistant', 'system']);
|
||||
expect(testMessage.content).toBeTruthy();
|
||||
});
|
||||
|
||||
tap.test('Interfaces: Provider options should be properly typed', async () => {
|
||||
// OpenAI options
|
||||
const openaiOptions: smartai.IOpenaiProviderOptions = {
|
||||
openaiToken: 'test-token',
|
||||
chatModel: 'gpt-5-mini',
|
||||
audioModel: 'tts-1-hd',
|
||||
visionModel: '04-mini',
|
||||
researchModel: 'o4-mini-deep-research-2025-06-26',
|
||||
enableWebSearch: true
|
||||
};
|
||||
|
||||
expect(openaiOptions).toBeInstanceOf(Object);
|
||||
expect(openaiOptions.openaiToken).toBeTruthy();
|
||||
|
||||
// Anthropic options
|
||||
const anthropicOptions: smartai.IAnthropicProviderOptions = {
|
||||
anthropicToken: 'test-token',
|
||||
enableWebSearch: true,
|
||||
searchDomainAllowList: ['example.com'],
|
||||
searchDomainBlockList: ['blocked.com']
|
||||
};
|
||||
|
||||
expect(anthropicOptions).toBeInstanceOf(Object);
|
||||
expect(anthropicOptions.anthropicToken).toBeTruthy();
|
||||
});
|
||||
|
||||
tap.test('Interfaces: Search depth values should be valid', async () => {
|
||||
const validDepths: smartai.ResearchOptions['searchDepth'][] = ['basic', 'advanced', 'deep'];
|
||||
|
||||
for (const depth of validDepths) {
|
||||
const options: smartai.ResearchOptions = {
|
||||
query: 'test',
|
||||
searchDepth: depth
|
||||
};
|
||||
expect(options.searchDepth).toBeOneOf(['basic', 'advanced', 'deep', undefined]);
|
||||
}
|
||||
});
|
||||
|
||||
tap.test('Interfaces: Optional properties should work correctly', async () => {
|
||||
// Minimal ResearchOptions
|
||||
const minimalOptions: smartai.ResearchOptions = {
|
||||
query: 'test query'
|
||||
};
|
||||
|
||||
expect(minimalOptions.query).toBeTruthy();
|
||||
expect(minimalOptions.searchDepth).toBeUndefined();
|
||||
expect(minimalOptions.maxSources).toBeUndefined();
|
||||
|
||||
// Minimal ChatOptions
|
||||
const minimalChat: smartai.ChatOptions = {
|
||||
systemMessage: 'system',
|
||||
userMessage: 'user',
|
||||
messageHistory: []
|
||||
};
|
||||
|
||||
expect(minimalChat.messageHistory).toBeArray();
|
||||
expect(minimalChat.messageHistory.length).toEqual(0);
|
||||
});
|
||||
|
||||
export default tap.start();
|
||||
@@ -9,14 +9,14 @@ import * as smartai from '../ts/index.js';
|
||||
|
||||
let testSmartai: smartai.SmartAi;
|
||||
|
||||
tap.test('should create a smartai instance', async () => {
|
||||
tap.test('OpenAI: should create a smartai instance with OpenAI provider', async () => {
|
||||
testSmartai = new smartai.SmartAi({
|
||||
openaiToken: await testQenv.getEnvVarOnDemand('OPENAI_TOKEN'),
|
||||
});
|
||||
await testSmartai.start();
|
||||
});
|
||||
|
||||
tap.test('should create chat response with openai', async () => {
|
||||
tap.test('OpenAI: should create chat response', async () => {
|
||||
const userMessage = 'How are you?';
|
||||
const response = await testSmartai.openaiProvider.chat({
|
||||
systemMessage: 'Hello',
|
||||
@@ -27,19 +27,21 @@ tap.test('should create chat response with openai', async () => {
|
||||
console.log(response.message);
|
||||
});
|
||||
|
||||
tap.test('should document a pdf', async () => {
|
||||
tap.test('OpenAI: should document a pdf', async () => {
|
||||
const pdfUrl = 'https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf';
|
||||
const pdfResponse = await smartrequest.getBinary(pdfUrl);
|
||||
const pdfResponse = await smartrequest.SmartRequest.create()
|
||||
.url(pdfUrl)
|
||||
.get();
|
||||
const result = await testSmartai.openaiProvider.document({
|
||||
systemMessage: 'Classify the document. Only the following answers are allowed: "invoice", "bank account statement", "contract", "other". The answer should only contain the keyword for machine use.',
|
||||
userMessage: "Classify the document.",
|
||||
messageHistory: [],
|
||||
pdfDocuments: [pdfResponse.body],
|
||||
pdfDocuments: [Buffer.from(await pdfResponse.arrayBuffer())],
|
||||
});
|
||||
console.log(result);
|
||||
});
|
||||
|
||||
tap.test('should recognize companies in a pdf', async () => {
|
||||
tap.test('OpenAI: should recognize companies in a pdf', async () => {
|
||||
const pdfBuffer = await smartfile.fs.toBuffer('./.nogit/demo_without_textlayer.pdf');
|
||||
const result = await testSmartai.openaiProvider.document({
|
||||
systemMessage: `
|
||||
@@ -76,7 +78,7 @@ tap.test('should recognize companies in a pdf', async () => {
|
||||
console.log(result);
|
||||
});
|
||||
|
||||
tap.test('should create audio response with openai', async () => {
|
||||
tap.test('OpenAI: should create audio response', async () => {
|
||||
// Call the audio method with a sample message.
|
||||
const audioStream = await testSmartai.openaiProvider.audio({
|
||||
message: 'This is a test of audio generation.',
|
||||
@@ -93,7 +95,7 @@ tap.test('should create audio response with openai', async () => {
|
||||
expect(audioBuffer.length).toBeGreaterThan(0);
|
||||
});
|
||||
|
||||
tap.test('should stop the smartai instance', async () => {
|
||||
tap.test('OpenAI: should stop the smartai instance', async () => {
|
||||
await testSmartai.stop();
|
||||
});
|
||||
|
||||
65
test/test.research.ts
Normal file
65
test/test.research.ts
Normal file
@@ -0,0 +1,65 @@
|
||||
import { tap, expect } from '@push.rocks/tapbundle';
|
||||
import * as smartai from '../ts/index.js';
|
||||
|
||||
// Test the research capabilities
|
||||
tap.test('OpenAI research method should exist', async () => {
|
||||
const openaiProvider = new smartai.OpenAiProvider({
|
||||
openaiToken: 'test-token'
|
||||
});
|
||||
|
||||
// Check that the research method exists
|
||||
expect(typeof openaiProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Anthropic research method should exist', async () => {
|
||||
const anthropicProvider = new smartai.AnthropicProvider({
|
||||
anthropicToken: 'test-token'
|
||||
});
|
||||
|
||||
// Check that the research method exists
|
||||
expect(typeof anthropicProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Research interfaces should be exported', async () => {
|
||||
// Check that the types are available (they won't be at runtime but TypeScript will check)
|
||||
const testResearchOptions: smartai.ResearchOptions = {
|
||||
query: 'test query',
|
||||
searchDepth: 'basic'
|
||||
};
|
||||
|
||||
expect(testResearchOptions).toBeInstanceOf(Object);
|
||||
expect(testResearchOptions.query).toEqual('test query');
|
||||
});
|
||||
|
||||
tap.test('Perplexity provider should have research method', async () => {
|
||||
const perplexityProvider = new smartai.PerplexityProvider({
|
||||
perplexityToken: 'test-token'
|
||||
});
|
||||
|
||||
// For Perplexity, we actually implemented it, so let's just check it exists
|
||||
expect(typeof perplexityProvider.research).toEqual('function');
|
||||
});
|
||||
|
||||
tap.test('Other providers should have research stubs', async () => {
|
||||
const groqProvider = new smartai.GroqProvider({
|
||||
groqToken: 'test-token'
|
||||
});
|
||||
|
||||
const ollamaProvider = new smartai.OllamaProvider({});
|
||||
|
||||
// Check that the research method exists and throws error
|
||||
expect(typeof groqProvider.research).toEqual('function');
|
||||
expect(typeof ollamaProvider.research).toEqual('function');
|
||||
|
||||
// Test that they throw errors when called
|
||||
let errorCaught = false;
|
||||
try {
|
||||
await groqProvider.research({ query: 'test' });
|
||||
} catch (error) {
|
||||
errorCaught = true;
|
||||
expect(error.message).toInclude('not yet supported');
|
||||
}
|
||||
expect(errorCaught).toBeTrue();
|
||||
});
|
||||
|
||||
export default tap.start();
|
||||
@@ -3,6 +3,6 @@
|
||||
*/
|
||||
export const commitinfo = {
|
||||
name: '@push.rocks/smartai',
|
||||
version: '0.5.4',
|
||||
version: '0.6.0',
|
||||
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.'
|
||||
}
|
||||
|
||||
@@ -1,3 +1,5 @@
|
||||
import * as plugins from './plugins.js';
|
||||
|
||||
/**
|
||||
* Message format for chat interactions
|
||||
*/
|
||||
@@ -23,22 +25,60 @@ export interface ChatResponse {
|
||||
message: string;
|
||||
}
|
||||
|
||||
/**
|
||||
* Options for research interactions
|
||||
*/
|
||||
export interface ResearchOptions {
|
||||
query: string;
|
||||
searchDepth?: 'basic' | 'advanced' | 'deep';
|
||||
maxSources?: number;
|
||||
includeWebSearch?: boolean;
|
||||
background?: boolean;
|
||||
}
|
||||
|
||||
/**
|
||||
* Response format for research interactions
|
||||
*/
|
||||
export interface ResearchResponse {
|
||||
answer: string;
|
||||
sources: Array<{
|
||||
url: string;
|
||||
title: string;
|
||||
snippet: string;
|
||||
}>;
|
||||
searchQueries?: string[];
|
||||
metadata?: any;
|
||||
}
|
||||
|
||||
/**
|
||||
* Abstract base class for multi-modal AI models.
|
||||
* Provides a common interface for different AI providers (OpenAI, Anthropic, Perplexity, Ollama)
|
||||
*/
|
||||
export abstract class MultiModalModel {
|
||||
/**
|
||||
* SmartPdf instance for document processing
|
||||
* Shared across all methods that need PDF functionality
|
||||
*/
|
||||
protected smartpdfInstance: plugins.smartpdf.SmartPdf;
|
||||
|
||||
/**
|
||||
* Initializes the model and any necessary resources
|
||||
* Should be called before using any other methods
|
||||
*/
|
||||
abstract start(): Promise<void>;
|
||||
public async start(): Promise<void> {
|
||||
this.smartpdfInstance = new plugins.smartpdf.SmartPdf();
|
||||
await this.smartpdfInstance.start();
|
||||
}
|
||||
|
||||
/**
|
||||
* Cleans up any resources used by the model
|
||||
* Should be called when the model is no longer needed
|
||||
*/
|
||||
abstract stop(): Promise<void>;
|
||||
public async stop(): Promise<void> {
|
||||
if (this.smartpdfInstance) {
|
||||
await this.smartpdfInstance.stop();
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Synchronous chat interaction with the model
|
||||
@@ -83,4 +123,12 @@ export abstract class MultiModalModel {
|
||||
pdfDocuments: Uint8Array[];
|
||||
messageHistory: ChatMessage[];
|
||||
}): Promise<{ message: any }>;
|
||||
|
||||
/**
|
||||
* Research and web search capabilities
|
||||
* @param optionsArg Options containing the research query and configuration
|
||||
* @returns Promise resolving to the research results with sources
|
||||
* @throws Error if the provider doesn't support research capabilities
|
||||
*/
|
||||
public abstract research(optionsArg: ResearchOptions): Promise<ResearchResponse>;
|
||||
}
|
||||
|
||||
@@ -91,7 +91,29 @@ export class SmartAi {
|
||||
}
|
||||
}
|
||||
|
||||
public async stop() {}
|
||||
public async stop() {
|
||||
if (this.openaiProvider) {
|
||||
await this.openaiProvider.stop();
|
||||
}
|
||||
if (this.anthropicProvider) {
|
||||
await this.anthropicProvider.stop();
|
||||
}
|
||||
if (this.perplexityProvider) {
|
||||
await this.perplexityProvider.stop();
|
||||
}
|
||||
if (this.groqProvider) {
|
||||
await this.groqProvider.stop();
|
||||
}
|
||||
if (this.xaiProvider) {
|
||||
await this.xaiProvider.stop();
|
||||
}
|
||||
if (this.ollamaProvider) {
|
||||
await this.ollamaProvider.stop();
|
||||
}
|
||||
if (this.exoProvider) {
|
||||
await this.exoProvider.stop();
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* create a new conversation
|
||||
|
||||
@@ -1,3 +1,9 @@
|
||||
export * from './classes.smartai.js';
|
||||
export * from './abstract.classes.multimodal.js';
|
||||
export * from './provider.openai.js';
|
||||
export * from './provider.anthropic.js';
|
||||
export * from './provider.perplexity.js';
|
||||
export * from './provider.groq.js';
|
||||
export * from './provider.ollama.js';
|
||||
export * from './provider.xai.js';
|
||||
export * from './provider.exo.js';
|
||||
|
||||
@@ -1,13 +1,16 @@
|
||||
import * as plugins from './plugins.js';
|
||||
import * as paths from './paths.js';
|
||||
import { MultiModalModel } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage, ResearchOptions, ResearchResponse } from './abstract.classes.multimodal.js';
|
||||
import type { ImageBlockParam, TextBlockParam } from '@anthropic-ai/sdk/resources/messages';
|
||||
|
||||
type ContentBlock = ImageBlockParam | TextBlockParam;
|
||||
|
||||
export interface IAnthropicProviderOptions {
|
||||
anthropicToken: string;
|
||||
enableWebSearch?: boolean;
|
||||
searchDomainAllowList?: string[];
|
||||
searchDomainBlockList?: string[];
|
||||
}
|
||||
|
||||
export class AnthropicProvider extends MultiModalModel {
|
||||
@@ -20,12 +23,15 @@ export class AnthropicProvider extends MultiModalModel {
|
||||
}
|
||||
|
||||
async start() {
|
||||
await super.start();
|
||||
this.anthropicApiClient = new plugins.anthropic.default({
|
||||
apiKey: this.options.anthropicToken,
|
||||
});
|
||||
}
|
||||
|
||||
async stop() {}
|
||||
async stop() {
|
||||
await super.stop();
|
||||
}
|
||||
|
||||
public async chatStream(input: ReadableStream<Uint8Array>): Promise<ReadableStream<string>> {
|
||||
// Create a TextDecoder to handle incoming chunks
|
||||
@@ -178,11 +184,10 @@ export class AnthropicProvider extends MultiModalModel {
|
||||
messageHistory: ChatMessage[];
|
||||
}): Promise<{ message: any }> {
|
||||
// Convert PDF documents to images using SmartPDF
|
||||
const smartpdfInstance = new plugins.smartpdf.SmartPdf();
|
||||
let documentImageBytesArray: Uint8Array[] = [];
|
||||
|
||||
for (const pdfDocument of optionsArg.pdfDocuments) {
|
||||
const documentImageArray = await smartpdfInstance.convertPDFToPngBytes(pdfDocument);
|
||||
const documentImageArray = await this.smartpdfInstance.convertPDFToPngBytes(pdfDocument);
|
||||
documentImageBytesArray = documentImageBytesArray.concat(documentImageArray);
|
||||
}
|
||||
|
||||
@@ -237,4 +242,121 @@ export class AnthropicProvider extends MultiModalModel {
|
||||
}
|
||||
};
|
||||
}
|
||||
|
||||
public async research(optionsArg: ResearchOptions): Promise<ResearchResponse> {
|
||||
// Prepare the messages for the research request
|
||||
const systemMessage = `You are a research assistant with web search capabilities.
|
||||
Provide comprehensive, well-researched answers with citations and sources.
|
||||
When searching the web, be thorough and cite your sources accurately.`;
|
||||
|
||||
try {
|
||||
// Build the tool configuration for web search
|
||||
const tools = this.options.enableWebSearch ? [
|
||||
{
|
||||
type: 'computer_20241022' as const,
|
||||
name: 'web_search',
|
||||
description: 'Search the web for current information',
|
||||
input_schema: {
|
||||
type: 'object' as const,
|
||||
properties: {
|
||||
query: {
|
||||
type: 'string',
|
||||
description: 'The search query'
|
||||
}
|
||||
},
|
||||
required: ['query']
|
||||
}
|
||||
}
|
||||
] : [];
|
||||
|
||||
// Configure the request based on search depth
|
||||
const maxTokens = optionsArg.searchDepth === 'deep' ? 8192 :
|
||||
optionsArg.searchDepth === 'advanced' ? 6144 : 4096;
|
||||
|
||||
// Create the research request
|
||||
const requestParams: any = {
|
||||
model: 'claude-3-opus-20240229',
|
||||
system: systemMessage,
|
||||
messages: [
|
||||
{
|
||||
role: 'user' as const,
|
||||
content: optionsArg.query
|
||||
}
|
||||
],
|
||||
max_tokens: maxTokens,
|
||||
temperature: 0.7
|
||||
};
|
||||
|
||||
// Add tools if web search is enabled
|
||||
if (tools.length > 0) {
|
||||
requestParams.tools = tools;
|
||||
requestParams.tool_choice = { type: 'auto' };
|
||||
}
|
||||
|
||||
// Execute the research request
|
||||
const result = await this.anthropicApiClient.messages.create(requestParams);
|
||||
|
||||
// Extract the answer from content blocks
|
||||
let answer = '';
|
||||
const sources: Array<{ url: string; title: string; snippet: string }> = [];
|
||||
const searchQueries: string[] = [];
|
||||
|
||||
// Process content blocks
|
||||
for (const block of result.content) {
|
||||
if ('text' in block) {
|
||||
answer += block.text;
|
||||
}
|
||||
}
|
||||
|
||||
// Parse sources from the answer (Claude includes citations in various formats)
|
||||
const urlRegex = /\[([^\]]+)\]\(([^)]+)\)/g;
|
||||
let match: RegExpExecArray | null;
|
||||
|
||||
while ((match = urlRegex.exec(answer)) !== null) {
|
||||
sources.push({
|
||||
title: match[1],
|
||||
url: match[2],
|
||||
snippet: ''
|
||||
});
|
||||
}
|
||||
|
||||
// Also look for plain URLs
|
||||
const plainUrlRegex = /https?:\/\/[^\s\)]+/g;
|
||||
const plainUrls = answer.match(plainUrlRegex) || [];
|
||||
|
||||
for (const url of plainUrls) {
|
||||
// Check if this URL is already in sources
|
||||
if (!sources.some(s => s.url === url)) {
|
||||
sources.push({
|
||||
title: new URL(url).hostname,
|
||||
url: url,
|
||||
snippet: ''
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
// Extract tool use information if available
|
||||
if ('tool_use' in result && Array.isArray(result.tool_use)) {
|
||||
for (const toolUse of result.tool_use) {
|
||||
if (toolUse.name === 'web_search' && toolUse.input?.query) {
|
||||
searchQueries.push(toolUse.input.query);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return {
|
||||
answer,
|
||||
sources,
|
||||
searchQueries: searchQueries.length > 0 ? searchQueries : undefined,
|
||||
metadata: {
|
||||
model: 'claude-3-opus-20240229',
|
||||
searchDepth: optionsArg.searchDepth || 'basic',
|
||||
tokensUsed: result.usage?.output_tokens
|
||||
}
|
||||
};
|
||||
} catch (error) {
|
||||
console.error('Anthropic research error:', error);
|
||||
throw new Error(`Failed to perform research: ${error.message}`);
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -1,7 +1,7 @@
|
||||
import * as plugins from './plugins.js';
|
||||
import * as paths from './paths.js';
|
||||
import { MultiModalModel } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage, ResearchOptions, ResearchResponse } from './abstract.classes.multimodal.js';
|
||||
import type { ChatCompletionMessageParam } from 'openai/resources/chat/completions';
|
||||
|
||||
export interface IExoProviderOptions {
|
||||
@@ -125,4 +125,8 @@ export class ExoProvider extends MultiModalModel {
|
||||
}): Promise<{ message: any }> {
|
||||
throw new Error('Document processing is not supported by Exo provider');
|
||||
}
|
||||
|
||||
public async research(optionsArg: ResearchOptions): Promise<ResearchResponse> {
|
||||
throw new Error('Research capabilities are not yet supported by Exo provider.');
|
||||
}
|
||||
}
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
import * as plugins from './plugins.js';
|
||||
import * as paths from './paths.js';
|
||||
import { MultiModalModel } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage, ResearchOptions, ResearchResponse } from './abstract.classes.multimodal.js';
|
||||
|
||||
export interface IGroqProviderOptions {
|
||||
groqToken: string;
|
||||
@@ -189,4 +189,8 @@ export class GroqProvider extends MultiModalModel {
|
||||
}): Promise<{ message: any }> {
|
||||
throw new Error('Document processing is not yet supported by Groq.');
|
||||
}
|
||||
|
||||
public async research(optionsArg: ResearchOptions): Promise<ResearchResponse> {
|
||||
throw new Error('Research capabilities are not yet supported by Groq provider.');
|
||||
}
|
||||
}
|
||||
@@ -1,7 +1,7 @@
|
||||
import * as plugins from './plugins.js';
|
||||
import * as paths from './paths.js';
|
||||
import { MultiModalModel } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage, ResearchOptions, ResearchResponse } from './abstract.classes.multimodal.js';
|
||||
|
||||
export interface IOllamaProviderOptions {
|
||||
baseUrl?: string;
|
||||
@@ -24,6 +24,7 @@ export class OllamaProvider extends MultiModalModel {
|
||||
}
|
||||
|
||||
async start() {
|
||||
await super.start();
|
||||
// Verify Ollama is running
|
||||
try {
|
||||
const response = await fetch(`${this.baseUrl}/api/tags`);
|
||||
@@ -35,7 +36,9 @@ export class OllamaProvider extends MultiModalModel {
|
||||
}
|
||||
}
|
||||
|
||||
async stop() {}
|
||||
async stop() {
|
||||
await super.stop();
|
||||
}
|
||||
|
||||
public async chatStream(input: ReadableStream<Uint8Array>): Promise<ReadableStream<string>> {
|
||||
// Create a TextDecoder to handle incoming chunks
|
||||
@@ -205,11 +208,10 @@ export class OllamaProvider extends MultiModalModel {
|
||||
messageHistory: ChatMessage[];
|
||||
}): Promise<{ message: any }> {
|
||||
// Convert PDF documents to images using SmartPDF
|
||||
const smartpdfInstance = new plugins.smartpdf.SmartPdf();
|
||||
let documentImageBytesArray: Uint8Array[] = [];
|
||||
|
||||
for (const pdfDocument of optionsArg.pdfDocuments) {
|
||||
const documentImageArray = await smartpdfInstance.convertPDFToPngBytes(pdfDocument);
|
||||
const documentImageArray = await this.smartpdfInstance.convertPDFToPngBytes(pdfDocument);
|
||||
documentImageBytesArray = documentImageBytesArray.concat(documentImageArray);
|
||||
}
|
||||
|
||||
@@ -249,4 +251,8 @@ export class OllamaProvider extends MultiModalModel {
|
||||
}
|
||||
};
|
||||
}
|
||||
|
||||
public async research(optionsArg: ResearchOptions): Promise<ResearchResponse> {
|
||||
throw new Error('Research capabilities are not yet supported by Ollama provider.');
|
||||
}
|
||||
}
|
||||
@@ -1,5 +1,6 @@
|
||||
import * as plugins from './plugins.js';
|
||||
import * as paths from './paths.js';
|
||||
import { Readable } from 'stream';
|
||||
|
||||
// Custom type definition for chat completion messages
|
||||
export type TChatCompletionRequestMessage = {
|
||||
@@ -8,19 +9,20 @@ export type TChatCompletionRequestMessage = {
|
||||
};
|
||||
|
||||
import { MultiModalModel } from './abstract.classes.multimodal.js';
|
||||
import type { ResearchOptions, ResearchResponse } from './abstract.classes.multimodal.js';
|
||||
|
||||
export interface IOpenaiProviderOptions {
|
||||
openaiToken: string;
|
||||
chatModel?: string;
|
||||
audioModel?: string;
|
||||
visionModel?: string;
|
||||
// Optionally add more model options (e.g., documentModel) if needed.
|
||||
researchModel?: string;
|
||||
enableWebSearch?: boolean;
|
||||
}
|
||||
|
||||
export class OpenAiProvider extends MultiModalModel {
|
||||
private options: IOpenaiProviderOptions;
|
||||
public openAiApiClient: plugins.openai.default;
|
||||
public smartpdfInstance: plugins.smartpdf.SmartPdf;
|
||||
|
||||
constructor(optionsArg: IOpenaiProviderOptions) {
|
||||
super();
|
||||
@@ -28,14 +30,16 @@ export class OpenAiProvider extends MultiModalModel {
|
||||
}
|
||||
|
||||
public async start() {
|
||||
await super.start();
|
||||
this.openAiApiClient = new plugins.openai.default({
|
||||
apiKey: this.options.openaiToken,
|
||||
dangerouslyAllowBrowser: true,
|
||||
});
|
||||
this.smartpdfInstance = new plugins.smartpdf.SmartPdf();
|
||||
}
|
||||
|
||||
public async stop() {}
|
||||
public async stop() {
|
||||
await super.stop();
|
||||
}
|
||||
|
||||
public async chatStream(input: ReadableStream<Uint8Array>): Promise<ReadableStream<string>> {
|
||||
// Create a TextDecoder to handle incoming chunks
|
||||
@@ -75,7 +79,7 @@ export class OpenAiProvider extends MultiModalModel {
|
||||
// If we have a complete message, send it to OpenAI
|
||||
if (currentMessage) {
|
||||
const messageToSend = { role: "user" as const, content: currentMessage.content };
|
||||
const chatModel = this.options.chatModel ?? 'o3-mini';
|
||||
const chatModel = this.options.chatModel ?? 'gpt-5-mini';
|
||||
const requestParams: any = {
|
||||
model: chatModel,
|
||||
messages: [messageToSend],
|
||||
@@ -121,7 +125,7 @@ export class OpenAiProvider extends MultiModalModel {
|
||||
content: string;
|
||||
}[];
|
||||
}) {
|
||||
const chatModel = this.options.chatModel ?? 'o3-mini';
|
||||
const chatModel = this.options.chatModel ?? 'gpt-5-mini';
|
||||
const requestParams: any = {
|
||||
model: chatModel,
|
||||
messages: [
|
||||
@@ -148,7 +152,8 @@ export class OpenAiProvider extends MultiModalModel {
|
||||
speed: 1,
|
||||
});
|
||||
const stream = result.body;
|
||||
done.resolve(stream);
|
||||
const nodeStream = Readable.fromWeb(stream as any);
|
||||
done.resolve(nodeStream);
|
||||
return done.promise;
|
||||
}
|
||||
|
||||
@@ -164,13 +169,10 @@ export class OpenAiProvider extends MultiModalModel {
|
||||
let pdfDocumentImageBytesArray: Uint8Array[] = [];
|
||||
|
||||
// Convert each PDF into one or more image byte arrays.
|
||||
const smartpdfInstance = new plugins.smartpdf.SmartPdf();
|
||||
await smartpdfInstance.start();
|
||||
for (const pdfDocument of optionsArg.pdfDocuments) {
|
||||
const documentImageArray = await smartpdfInstance.convertPDFToPngBytes(pdfDocument);
|
||||
const documentImageArray = await this.smartpdfInstance.convertPDFToPngBytes(pdfDocument);
|
||||
pdfDocumentImageBytesArray = pdfDocumentImageBytesArray.concat(documentImageArray);
|
||||
}
|
||||
await smartpdfInstance.stop();
|
||||
|
||||
console.log(`image smartfile array`);
|
||||
console.log(pdfDocumentImageBytesArray.map((smartfile) => smartfile.length));
|
||||
@@ -184,7 +186,7 @@ export class OpenAiProvider extends MultiModalModel {
|
||||
},
|
||||
}));
|
||||
|
||||
const chatModel = this.options.chatModel ?? 'o4-mini';
|
||||
const chatModel = this.options.chatModel ?? 'gpt-5-mini';
|
||||
const requestParams: any = {
|
||||
model: chatModel,
|
||||
messages: [
|
||||
@@ -229,4 +231,111 @@ export class OpenAiProvider extends MultiModalModel {
|
||||
const result = await this.openAiApiClient.chat.completions.create(requestParams);
|
||||
return result.choices[0].message.content || '';
|
||||
}
|
||||
|
||||
public async research(optionsArg: ResearchOptions): Promise<ResearchResponse> {
|
||||
// Determine which model to use based on search depth
|
||||
let model: string;
|
||||
if (optionsArg.searchDepth === 'deep') {
|
||||
model = this.options.researchModel || 'o4-mini-deep-research-2025-06-26';
|
||||
} else {
|
||||
model = this.options.chatModel || 'gpt-5-mini';
|
||||
}
|
||||
|
||||
// Prepare the request parameters
|
||||
const requestParams: any = {
|
||||
model,
|
||||
messages: [
|
||||
{
|
||||
role: 'system',
|
||||
content: 'You are a research assistant. Provide comprehensive answers with citations and sources when available.'
|
||||
},
|
||||
{
|
||||
role: 'user',
|
||||
content: optionsArg.query
|
||||
}
|
||||
],
|
||||
temperature: 0.7
|
||||
};
|
||||
|
||||
// Add web search tools if requested
|
||||
if (optionsArg.includeWebSearch || optionsArg.searchDepth === 'deep') {
|
||||
requestParams.tools = [
|
||||
{
|
||||
type: 'function',
|
||||
function: {
|
||||
name: 'web_search',
|
||||
description: 'Search the web for information',
|
||||
parameters: {
|
||||
type: 'object',
|
||||
properties: {
|
||||
query: {
|
||||
type: 'string',
|
||||
description: 'The search query'
|
||||
}
|
||||
},
|
||||
required: ['query']
|
||||
}
|
||||
}
|
||||
}
|
||||
];
|
||||
requestParams.tool_choice = 'auto';
|
||||
}
|
||||
|
||||
// Add background flag for deep research
|
||||
if (optionsArg.background && optionsArg.searchDepth === 'deep') {
|
||||
requestParams.background = true;
|
||||
}
|
||||
|
||||
try {
|
||||
// Execute the research request
|
||||
const result = await this.openAiApiClient.chat.completions.create(requestParams);
|
||||
|
||||
// Extract the answer
|
||||
const answer = result.choices[0].message.content || '';
|
||||
|
||||
// Parse sources from the response (OpenAI often includes URLs in markdown format)
|
||||
const sources: Array<{ url: string; title: string; snippet: string }> = [];
|
||||
const urlRegex = /\[([^\]]+)\]\(([^)]+)\)/g;
|
||||
let match: RegExpExecArray | null;
|
||||
|
||||
while ((match = urlRegex.exec(answer)) !== null) {
|
||||
sources.push({
|
||||
title: match[1],
|
||||
url: match[2],
|
||||
snippet: '' // OpenAI doesn't provide snippets in standard responses
|
||||
});
|
||||
}
|
||||
|
||||
// Extract search queries if tools were used
|
||||
const searchQueries: string[] = [];
|
||||
if (result.choices[0].message.tool_calls) {
|
||||
for (const toolCall of result.choices[0].message.tool_calls) {
|
||||
if ('function' in toolCall && toolCall.function.name === 'web_search') {
|
||||
try {
|
||||
const args = JSON.parse(toolCall.function.arguments);
|
||||
if (args.query) {
|
||||
searchQueries.push(args.query);
|
||||
}
|
||||
} catch (e) {
|
||||
// Ignore parsing errors
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return {
|
||||
answer,
|
||||
sources,
|
||||
searchQueries: searchQueries.length > 0 ? searchQueries : undefined,
|
||||
metadata: {
|
||||
model,
|
||||
searchDepth: optionsArg.searchDepth || 'basic',
|
||||
tokensUsed: result.usage?.total_tokens
|
||||
}
|
||||
};
|
||||
} catch (error) {
|
||||
console.error('Research API error:', error);
|
||||
throw new Error(`Failed to perform research: ${error.message}`);
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -1,7 +1,7 @@
|
||||
import * as plugins from './plugins.js';
|
||||
import * as paths from './paths.js';
|
||||
import { MultiModalModel } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage, ResearchOptions, ResearchResponse } from './abstract.classes.multimodal.js';
|
||||
|
||||
export interface IPerplexityProviderOptions {
|
||||
perplexityToken: string;
|
||||
@@ -168,4 +168,69 @@ export class PerplexityProvider extends MultiModalModel {
|
||||
}): Promise<{ message: any }> {
|
||||
throw new Error('Document processing is not supported by Perplexity.');
|
||||
}
|
||||
|
||||
public async research(optionsArg: ResearchOptions): Promise<ResearchResponse> {
|
||||
// Perplexity has Sonar models that are optimized for search
|
||||
// sonar models: sonar, sonar-pro
|
||||
const model = optionsArg.searchDepth === 'deep' ? 'sonar-pro' : 'sonar';
|
||||
|
||||
try {
|
||||
const response = await fetch('https://api.perplexity.ai/chat/completions', {
|
||||
method: 'POST',
|
||||
headers: {
|
||||
'Authorization': `Bearer ${this.options.perplexityToken}`,
|
||||
'Content-Type': 'application/json',
|
||||
},
|
||||
body: JSON.stringify({
|
||||
model,
|
||||
messages: [
|
||||
{
|
||||
role: 'system',
|
||||
content: 'You are a helpful research assistant. Provide accurate information with sources.'
|
||||
},
|
||||
{
|
||||
role: 'user',
|
||||
content: optionsArg.query
|
||||
}
|
||||
],
|
||||
temperature: 0.7,
|
||||
max_tokens: 4000
|
||||
}),
|
||||
});
|
||||
|
||||
if (!response.ok) {
|
||||
throw new Error(`Perplexity API error: ${response.statusText}`);
|
||||
}
|
||||
|
||||
const result = await response.json();
|
||||
const answer = result.choices[0].message.content;
|
||||
|
||||
// Parse citations from the response
|
||||
const sources: Array<{ url: string; title: string; snippet: string }> = [];
|
||||
|
||||
// Perplexity includes citations in the format [1], [2], etc. with sources listed
|
||||
// This is a simplified parser - could be enhanced based on actual Perplexity response format
|
||||
if (result.citations) {
|
||||
for (const citation of result.citations) {
|
||||
sources.push({
|
||||
url: citation.url || '',
|
||||
title: citation.title || '',
|
||||
snippet: citation.snippet || ''
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
return {
|
||||
answer,
|
||||
sources,
|
||||
metadata: {
|
||||
model,
|
||||
searchDepth: optionsArg.searchDepth || 'basic'
|
||||
}
|
||||
};
|
||||
} catch (error) {
|
||||
console.error('Perplexity research error:', error);
|
||||
throw new Error(`Failed to perform research: ${error.message}`);
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -1,7 +1,7 @@
|
||||
import * as plugins from './plugins.js';
|
||||
import * as paths from './paths.js';
|
||||
import { MultiModalModel } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage } from './abstract.classes.multimodal.js';
|
||||
import type { ChatOptions, ChatResponse, ChatMessage, ResearchOptions, ResearchResponse } from './abstract.classes.multimodal.js';
|
||||
import type { ChatCompletionMessageParam } from 'openai/resources/chat/completions';
|
||||
|
||||
export interface IXAIProviderOptions {
|
||||
@@ -11,7 +11,6 @@ export interface IXAIProviderOptions {
|
||||
export class XAIProvider extends MultiModalModel {
|
||||
private options: IXAIProviderOptions;
|
||||
public openAiApiClient: plugins.openai.default;
|
||||
public smartpdfInstance: plugins.smartpdf.SmartPdf;
|
||||
|
||||
constructor(optionsArg: IXAIProviderOptions) {
|
||||
super();
|
||||
@@ -19,14 +18,16 @@ export class XAIProvider extends MultiModalModel {
|
||||
}
|
||||
|
||||
public async start() {
|
||||
await super.start();
|
||||
this.openAiApiClient = new plugins.openai.default({
|
||||
apiKey: this.options.xaiToken,
|
||||
baseURL: 'https://api.x.ai/v1',
|
||||
});
|
||||
this.smartpdfInstance = new plugins.smartpdf.SmartPdf();
|
||||
}
|
||||
|
||||
public async stop() {}
|
||||
public async stop() {
|
||||
await super.stop();
|
||||
}
|
||||
|
||||
public async chatStream(input: ReadableStream<Uint8Array>): Promise<ReadableStream<string>> {
|
||||
// Create a TextDecoder to handle incoming chunks
|
||||
@@ -180,4 +181,8 @@ export class XAIProvider extends MultiModalModel {
|
||||
message: completion.choices[0]?.message?.content || ''
|
||||
};
|
||||
}
|
||||
|
||||
public async research(optionsArg: ResearchOptions): Promise<ResearchResponse> {
|
||||
throw new Error('Research capabilities are not yet supported by xAI provider.');
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user