Pi integrates MCP protocol while reducing token overhead
Pi, the coding agent acquired by Earendil, now supports the MCP protocol used for browser automation—but avoids the massive token cost that comes with exposing all of MCP’s tools at once. Instead, Pi puts its Codemode system between the language model and these tools, loading tool details only when needed instead of upfront. This brings MCP’s integration down from consuming over 18,000 tokens to something far leaner.
- MCP protocol previously cost 18,000 tokens per server
- Pi uses Codemode to call tools only on demand
- Most tool definitions stay out of the model’s context
- Earendil says workaround makes Pi more flexible
- Developers can still expose tools directly if needed
Sources covering this
More in AI
OpenAI trials visual ads in ChatGPT's image generator
OpenAI will start testing visual ads in ChatGPT this month, showing paid product images while users generate AI images.
Reflection launches Beam, a large open AI model with cheaper compute
Reflection, a well-funded US AI startup, released Beam, a massive open-weight AI model it claims rivals China’s best open models in…
OpenAI speeds up GPT-6 Astra and Sol models
OpenAI made its GPT-6 Astra and Sol models about 50% faster, now generating around 50 tokens per second instead of 30.
Professor swaps coding tests for AI-guided interviews
A university professor has replaced traditional coding exams with interviews that test how students use AI tools like Gemini in Google…