recipes
Docs
GitHub
Community Recipes
Spark Arena
← Back to recipes
qwen3.5-122b-fp8
vllm
ray
@eugr
vLLM serving Qwen3.5-122B-FP8
Quick Info
Model
Qwen/Qwen3.5-122B-A10B-FP8
Runtime
vllm (ray)
Tensor Parallel
2
Nodes
2+
Dtype
fp8
Container
vllm-node
Command Builder
Tensor Parallel (--tp)
GPU Memory (--gpu-mem)
Port (--port)
Served Model Name
sparkrun run @eugr/qwen3.5-122b-fp8
View Recipe YAML