fix: pin models to version, except for gpt-oss
This commit is contained in:
@@ -8,4 +8,5 @@ kv-cache-dtype: fp8
|
||||
enforce-eager: false
|
||||
enable-prefix-caching: true
|
||||
enable-chunked-prefill: true
|
||||
vllm-release: v2.22.5
|
||||
speculative-config: '{"model":"RedHatAI/gemma-4-31B-it-speculator.eagle3","method":"eagle3","num_speculative_tokens":3}'
|
||||
|
||||
Reference in New Issue
Block a user