There’s a bug open for this: [Bug]: Qwen Coder Next prefix caching · Issue #34361 · vllm-project/vllm · GitHub
This version is stable (prior to the change)
./build-and-copy.sh -t vllm-node-20260209 -c --vllm-ref 13397841ab469cecf1ed425c3f52a9ffc38139b5
more context at: Introducing the Spark Arena - #17 by eugr