2025-11-14 19:39:24,670] [3917] [281451364479328] [llm] [INFO] [cache_manager.py-79] : Before prefill warmup, peak mem: 16.85G, total_mem: 21.02G
[2025-11-14 19:39:24,670] [3917] [281451364479328] [llm] [INFO] [generator.py-530] : warmup params: max_prefill_tokens=256, max_seq_len=256, max_input_len=256, max_iter_times=1024
[2025-11-14 19:39:24,670] [3917] [281451364479328] [llm] [WARN] [generator.py-588] : The `max_prefill_tokens` exceeds what the npu mem can hold.
[2025-11-14 19:39:24,671] [3917] [281451364479328] [llm] [ERROR] [logging.py-55] : Warmup failed. This issue could be caused by setting both `max_prefill_tokens` and `max_input_length` to very large values, or setting insufficient `npu_mem`. Reducing either `max_prefill_tokens` or `max_input_length` may help resolve this issue. If `npu_mem` is -1, try to increase the environment value `NPU_MEMORY_FRACTION` or `npu_mem` in configuration directly. Increase `world_size` can be another choice.