{"schema":"postcutoff/event@1","as_of":"2026-11-09T21:13:00+01:00","url":"https://postcutoff.com/e/2026-11-08-llama-modchip-deepseek-v4-flash-cpu-64gb/","md":"https://postcutoff.com/e/2026-11-08-llama-modchip-deepseek-v4-flash-cpu-64gb/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-11-08-llama-modchip-deepseek-v4-flash-cpu-64gb","date":"2026-11-08","date_precision":"day","short_title":"llama-modchip: open-source project runs 284B DeepSeek-V4-Flash on a 64 GB CPU-only machine at 1.71 bits per weight","deck":null,"takeaway":"It reached the Hacker News front page on Nov 8 (432 points).","category":"open-source","category_label":"Open source","importance":2,"confidence":"medium","status":{"key":"partly","labels":["Partly confirmed"]},"sources":[{"n":1,"title":"GitHub: synchopate/llama-modchip","url":"https://github.com/synchopate/llama-modchip","type":"code","group":"primary","domain":"github.com"},{"n":2,"title":"Hacker News discussion","url":"https://news.ycombinator.com/item?id=50411961","type":"discussion","group":"reactions","domain":"news.ycombinator.com"}],"official":1,"filed":"2026-11-09","updated":"2026-11-09","orgs":["synchopate"],"title":"llama-modchip: open-source project runs 284B DeepSeek-V4-Flash on a 64 GB CPU-only machine at 1.71 bits per weight","summary":"llama-modchip, an Apache-2.0 fork-style CPU inference project on GitHub (repo created Oct 29, 2026), claims to run DeepSeek-V4-Flash (284B MoE, 13B active) in under 59 GiB of RAM with no GPU, at a 1.71-bit GGUF quantization, 256K context and 9–15 tokens/s decode on a 16-core Zen 5. It reached the Hacker News front page on Nov 8 (432 points). Figures are the author's own; quality loss at 1.71 bpw is not independently measured.","key_facts":["Claims (README v2.4.0): 284.33B total / 13.17B active; 56.56 GiB GGUF at 1.71 bpw; peak 57.61 GiB RSS at 256K context; 44.4 tok/s prefill and 9.1–14.7 tok/s decode on CPU; 30B MoEs at 25–111 tok/s","~650 GitHub stars by Nov 9; HN 432 points"],"key_numbers":[],"tags":["local-inference","quantization","cpu","deepseek","open-weights"],"science":null,"body_md":"","disputed":[],"related":[],"people":[],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-11-09","type":"filed","text":"Created (news agent)"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-11-09","run":null,"sources_read":null,"updated":"2026-11-09","human_review":null,"version":{"date":"2026-11-09"}},"gaps":[{"model_id":"gpt-6-astra","name":"GPT-6 Astra","cutoff":"2026-04","days_after":192,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":131,"in_training_data":false},{"model_id":"gemini-4-argon","name":"Gemini 4 Argon","cutoff":"2026-03","days_after":222,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":161,"in_training_data":false}],"short_url":null}