▲ 167 today·628 stars·Python·HN ×1New · 12d old

synchopate/llama-modchip

Repo · synchopate/llama-modchip

InferenceLLMsRAG & Memory

What is llama-modchip?

🐋 Run 284B DeepSeek-V4-Flash (<59GiB RAM, 44.4 tok/s pp, 9.1-14.7 tok/s) & 30B MoEs (25-111 tok/s) on a pure 64GB CPU at 256K ctx (0 GPUs). SwiGLU-H2, Stockfish-PVS L3 & 13ms Claude Code restore.

Try it in 60 seconds

No one-line install. The README has the setup steps.

Apache-2.0 licenseUpdated todayPython19 open issues

New projects are not vetted. Skim the code, and run it in a fresh folder or a container before giving it your files or keys.

Open on GitHub → Website / demo

The morning email · free

Get tomorrow's biggest AI launches before everyone else.

One short email each morning: what launched, why it matters, how to try it. Five minutes, no spam, unsubscribe any time.

Email addressSubscribe

The story so far

Early Signal first spotted synchopate/llama-modchip on November 8, 2026, when it had 628 GitHub stars. It now has 628. On its busiest day it gained 167 stars. The project is 12 days old and written mainly in Python. It is published under the Apache-2.0 license and has 43 forks. It has been discussed in 1 Hacker News thread, the top one with 429 points. Early Signal files it under Inference, LLMs, RAG & Memory.

Questions about llama-modchip

Is llama-modchip free and open source?

Yes. synchopate/llama-modchip is open source under the Apache-2.0 license, and the code is on GitHub at https://github.com/synchopate/llama-modchip.