zijie-tian/nano-vllm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[docs] refactor CLAUDE.md.

Browse Source
This commit is contained in:
Zijie Tian
2025-12-15 21:43:33 +08:00
parent dc7807a211
commit 8df0c7517b
2 changed files with 70 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
.claude/rules/commands.md
View File

@@ -21,6 +21,6 @@ python bench_offload.py # CPU offload benchmark
- `max_num_batched_tokens`: 16384
- `max_num_seqs`: 512
- `kvcache_block_size`: 256
- `kvcache_block_size`: 4096
- `gpu_memory_utilization`: 0.9
- `enforce_eager`: False (enables CUDA graphs)