vllm offline inference issue
#161
by
catalin-hanga
- opened
When I try to run the example for vllm offline inference from the homepage (https://github.com/openai/gpt-oss?tab=readme-ov-file#vllm), the following line
outputs = llm.generate(prompt_token_ids=[prefill_ids], sampling_params=sampling)
causes this error message:
TypeError: LLM.generate() got an unexpected keyword argument 'prompt_token_ids'
Which version of vllm is used in this example ?
catalin-hanga
changed discussion status to
closed