getcompanion-ai/co-mono
1
0
Fork 0
mirror of https://github.com/getcompanion-ai/co-mono.git synced 2026-04-15 17:00:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
co-mono/packages/ai
History
Mario Zechner 3e1422d3d7 feat(ai): Add proper thinking support for Gemini 2.5 models 
- Added thinkingConfig with includeThoughts and thinkingBudget support
- Use part.thought boolean flag to detect thinking content per API docs
- Capture and preserve thought signatures for multi-turn function calling
- Added supportsThinking() check for Gemini 2.5 series models
- Updated example to demonstrate thinking configuration
- Handle SDK type limitations with proper type assertions
2025-08-25 10:26:23 +02:00
..
docs refactor(ai): Add completion signal to onText/onThinking callbacks 2025-08-24 20:33:26 +02:00
src feat(ai): Add proper thinking support for Gemini 2.5 models 2025-08-25 10:26:23 +02:00
test/examples feat(ai): Add proper thinking support for Gemini 2.5 models 2025-08-25 10:26:23 +02:00
package.json feat(ai): Implement Gemini provider with streaming and tool support 2025-08-24 20:41:32 +02:00
README.md feat(ai): Create unified AI package with OpenAI, Anthropic, and Gemini support 2025-08-17 20:18:45 +02:00
tsconfig.build.json feat(ai): Create unified AI package with OpenAI, Anthropic, and Gemini support 2025-08-17 20:18:45 +02:00

README.md

@mariozechner/ai

Unified API for OpenAI, Anthropic, and Google Gemini LLM providers. This package provides a common interface for working with multiple LLM providers, handling their differences transparently while exposing a consistent, minimal API.

Features (Planned)

  • Unified Interface: Single API for OpenAI, Anthropic, and Google Gemini
  • Streaming Support: Real-time response streaming with delta events
  • Tool Calling: Consistent tool/function calling across providers
  • Reasoning/Thinking: Support for reasoning tokens where available
  • Session Management: Serializable conversation state across providers
  • Token Tracking: Unified token counting (input, output, cached, reasoning)
  • Interrupt Handling: Graceful cancellation of requests
  • Provider Detection: Automatic configuration based on endpoint
  • Caching Support: Provider-specific caching strategies

Installation

npm install @mariozechner/ai

Quick Start (Coming Soon)

import { createClient } from '@mariozechner/ai';

// Automatically detects provider from configuration
const client = createClient({
  provider: 'openai',
  apiKey: process.env.OPENAI_API_KEY,
  model: 'gpt-4'
});

// Same API works for all providers
const response = await client.complete({
  messages: [
    { role: 'user', content: 'Hello!' }
  ],
  stream: true
});

for await (const event of response) {
  if (event.type === 'content') {
    process.stdout.write(event.text);
  }
}

Supported Providers

  • OpenAI: GPT-3.5, GPT-4, o1, o3 models
  • Anthropic: Claude models via native SDK
  • Google Gemini: Gemini models with thinking support

Development

This package is part of the pi monorepo. See the main README for development instructions.

License

MIT