gahow/aituner
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Pin 27B A/B to GPUs 2-7 (route around leaked GPU0/1 memory)

Browse Source 
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Gahow Wang
2026-06-15 23:01:22 +08:00
parent 2fcaf80450
commit b1b74318f6
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
configs/examples/dash0_qwen27b_tp_ab.json
View File

@@ -48,7 +48,8 @@
"VLLM_DP_MASTER_PORT": "9528", "VLLM_DP_MASTER_PORT": "9528",
"VLLM_RESPONSE_TIMEOUT": "300", "VLLM_RESPONSE_TIMEOUT": "300",
"VLLM_LOG_REQ_KV_LENS": "1", "VLLM_LOG_REQ_KV_LENS": "1",
"DS_LLM_GRACEFUL_SHUTDOWN_KEEP_SECONDS": "600" "DS_LLM_GRACEFUL_SHUTDOWN_KEEP_SECONDS": "600",
"CUDA_VISIBLE_DEVICES": "2,3,4,5,6,7"
}, },
"base_flags": { "base_flags": {
"host": "127.0.0.1", "host": "127.0.0.1",