Zer0Fit: Local MCP Wrapper for TabFM/TimesFM
Porespellar · reddit · 2026-07-12
The author created a locally available MCP service for Google's newly released TabFM and TimesFM, named Zer0Fit. The goal is to enable local LLMs to perform zero-shot ML tasks like tabular classification, regression, and time-series forecasting directly via a chat interface.
Project highlights:
- Runs as a single Docker container with dynamic model loading/unloading; TTL is set to 5 minutes to save VRAM.
- Currently supports CSV, with future plans for XLS/XLSX/JSON/JSONL.
- Supports only CUDA / NVIDIA environments. The author notes it works on devices like DGX Spark, 3090, and H100, requiring at least 16GB VRAM.
- Test datasets include Iris, California Housing, and Airline Passengers. The author achieved 94.7% accuracy on Iris and an R2 of 0.91 for regression.
- Provides an Open WebUI Skill.md, along with integration support for Claude Code and Codex CLI.
The author believes these foundation models effectively "GenAI-ify" parts of traditional ML, making them accessible for regular users to "try out" model capabilities without manual hyperparameter tuning.
Related event: Zer0Fit: Local MCP Server for TabFM and TimesFM(2 posts)→
More from coding & agent
- Agent Field Kit launches $29 security runbooks for production AI agents — alvelda · 2026-10-27
- Dev mixes Claude specs with abliterated GLM to build SS7 threat detector — l4rz · 2026-10-27
- Cloudflare Durable Objects validated as top architecture for AI agent WebSocket servers — tristanbob · 2026-10-27
- GitHub Next's Chopin prototype explores real-time multiplayer agentic planning — max__drake · 2026-10-27
- Veynod: An Open-Source Shared Workspace Where AI Agents Collaborate Like a Team — TKOWNS · 2026-10-27
- TMLR paper: enforced thinking doesn't reliably make LLM agents disclose key info to users — SharonYixuanLi · 2026-10-27