anyone knows how to limit thinking length in vllm openai server? it seems like need to adit the logits processor.
anyone knows how to limit thinking length in vllm openai server? it seems like need to adit the logits processor.