Files
worker-vllm/configs/gemma/gemma4_31b_it.yaml
T

12 lines
351 B
YAML

model: google/gemma-4-31b-it
gpu-memory-utilization: 0.95
max-model-len: 8192
dtype: auto
trust-remote-code: true
quantization: fp8
kv-cache-dtype: fp8
enforce-eager: false
enable-prefix-caching: true
enable-chunked-prefill: true
speculative-config: '{"model":"RedHatAI/gemma-4-31B-it-speculator.eagle3","method":"eagle3","num_speculative_tokens":3}'