composte-hermes.yml aktualisiert

This commit is contained in:
Marc Blatter 2026-07-26 17:58:52 +00:00
parent d080f54087
commit bbe88f9203
1 changed files with 1 additions and 1 deletions

View File

@ -26,7 +26,7 @@ services:
- --quantization=awq - --quantization=awq
- --tensor-parallel-size=2 - --tensor-parallel-size=2
- --gpu-memory-utilization=0.86 - --gpu-memory-utilization=0.86
- --max-model-len=65536 - --max-model-len=60000
- --hf-overrides={"rope_scaling":{"rope_type":"yarn","factor":2.0,"original_max_position_embeddings":32768}} - --hf-overrides={"rope_scaling":{"rope_type":"yarn","factor":2.0,"original_max_position_embeddings":32768}}
- --max-num-seqs=4 - --max-num-seqs=4
- --served-model-name=qwen3-14b-awq - --served-model-name=qwen3-14b-awq