build1 publisher
Pointing the Ray head and workers at vLLM's own image removes the numpy 2.0 ABI crash
A dev.to walkthrough of a 600GB NVFP4 model on discounted 8xH100 spot nodes traces the two crashes that arrive before the first prompt to a pip resolver replacing numpy and a KV cache sized for a million tokens.
Publishers:dev.to
Reality
- Evidence26
- Adoption14
- Hype gap+34
- Incentives
- Insufficient
- Confidence28