Swapping AI SDK for pi-ai as the LLM provider layer
Melvin Vivas · X post · 2026-08-16 · Open on X
Topics: LLM Fundamentals, AI Agents, Tool Use & MCP · Level: intermediate
Summary
The creator replaced the AI SDK in his own AI library, which connects to multiple AI providers, with pi-ai from Pi (@pidotdev). It's an example of choosing a provider-abstraction layer for calling different LLM APIs through one interface.
Key points
- Use one library to talk to many LLM providers instead of coding against each API
- The creator moved his library from AI SDK to pi-ai
- He found pi-ai the best fit for interfacing with AI providers
Resources mentioned
- Pi · tool · x.com · free
A customizable coding agent that can be extended through its Extensions API and connected to several model providers.
Also in: 100-Iteration Autonomous Coding Loop with Local Qwen on an RTX 3090 (Melvin Vivas on X · notes), Pi Agent: A Highly Customizable Agent with Durable Sessions (Melvin Vivas on X · notes), Prompting an Autonomous Agent Loop: Be Specific, Show Visible Progress (Melvin Vivas on X · notes), Running an Overnight 100-Iteration Agent Loop with a Local Qwen Model (Melvin Vivas on X · notes) and 54 more - pi-ai · tool · github.com · free · open in a browser to verify
A library from @pidotdev (pi.dev) that Coworker uses to connect to LLM providers.
Also in: Coworker v0.7.0: Open-Source Multi-Bot Workspace with Primary Bot and Web Search (Melvin Vivas on X · notes) - Vercel AI SDK · tool · ai-sdk.dev · free · recommended by both Bashiri Smith & Melvin Vivas
TypeScript SDK for adding AI and agent features to web products.
More in LLM Fundamentals
- Qwen 3.8 27B Matches GPT 5.6 Luna (max) on Artificial Analysis
- Run LFM2.5-2.6B locally with llama-cpp-python in Colab
- GLM-5.3 free weekend trial announcement
- Open Models Aren't Always Local Models
- Using LFM2.5-VL-3B's Vision Capabilities (Liquid AI Guide)
- Cohere's North Micro Vision: a small open-source vision model for documents