Issue evidence · Serve · vllm-project/vllm
GitHub issue: https://github.com/vllm-project/vllm/issues/61069
Not run
The engine dies roughly once a day after about 2,000 sequential short requests on a GB10.
Nothing was run: GB10 (sm_121) is not a GPU Badgr's providers rent.
A funded run would need GB10 hardware; a different GPU would not test the same kernel path.
Checked 2026-10-11 in Badgr’s local development environment.
← All issue evidence