Files
deepseek-v4-quant/memory/2026-04-15.md
biondizzle f63eed5cfd Purge INT4 references — expert weights are FP4 (E2M1), not INT4
All docs and scripts updated. Historical memory entries annotated.
2026-05-08 02:33:46 +00:00

10 lines
363 B
Markdown

# 2026-04-15
## Updated vLLM with LMCache Dockerfile
Mike asked me to modify the Dockerfile at `/home/openclaw/dev/vllm-with-lmcache/Dockerfile` to:
- Clone from `https://github.com/neuralwatt/LMCache.git` instead of original LMCache repo
- Checkout branch `fix/mla-multi-group-kv-cache`
- Build from that branch
Successfully updated the file with the changes.