vllm tweaks

This commit is contained in:
2026-05-17 07:10:16 +00:00
parent 48e4cb625d
commit 34c43958d0

View File

@@ -19,7 +19,12 @@ services:
- --compilation-config
#- '{"cudagraph_mode": "FULL_DECODE_ONLY", "custom_ops": ["all"], "cudagraph_capture_sizes": [1, 2, 4, 8], "max_cudagraph_capture_size": 8}'
- '{"cudagraph_mode":"FULL_AND_PIECEWISE", "custom_ops":["all"]}'
- --moe-backend deep_gemm_mega_moe
- --tokenizer-mode=deepseek_v4
#- --attention_config.use_fp4_indexer_cache=True
- --tool-call-parser deepseek_v4
- --enable-auto-tool-choice
- --reasoning-parser deepseek_v4
- --host=0.0.0.0
- --port=8000
deploy: