biondizzle/nvfp4-megamoe-kernel
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

crap shoot

Browse Source
This commit is contained in:
biondizzle
2026-05-17 16:25:38 +00:00
parent 7c16f3cb46
commit c03438fc4e
1 changed files with 3 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
docker-compose.yml
View File

@@ -17,14 +17,15 @@ services:
- --tensor-parallel-size=8
#- --enforce-eager
- --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"]}'
- '{"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
- --gpu_memory_utilization=0.9
- --host=0.0.0.0
- --port=8000
deploy: