InferenceMAX
InferenceMAX copied to clipboard
AMD needs to use upstream vLLM images instead of fork
For instance