执行 FunASR/examples/industrial_data_pretraining/fun_asr_nano 目录下的 serve_realtime_ws.py,执行命令如下:
2026-05-25 10:57:39,765 [INFO] Loading ASR (vLLM): FunAudioLLM/Fun-ASR-Nano-2512
2026-05-25 10:57:39,875 [INFO] HTTP Request: GET https://huggingface.co/api/models/FunAudioLLM/Fun-ASR-Nano-2512/revision/main "HTTP/1.1 200 OK"
Downloading (incomplete total...): 0.00B [00:00, ?B/s] 2026-05-25 10:57:39,958 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/.gitattributes "HTTP/1.1 307 Temporary Redirect" | 0/21 [00:00<?, ?it/s]
2026-05-25 10:57:39,961 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/Qwen3-0.6B/merges.txt "HTTP/1.1 307 Temporary Redirect"
Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
2026-05-25 10:57:39,962 [WARNING] Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
2026-05-25 10:57:39,963 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/.gitattributes "HTTP/1.1 200 OK"
2026-05-25 10:57:39,964 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/Qwen3-0.6B/generation_config.json "HTTP/1.1 302 Found"
2026-05-25 10:57:39,965 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/Qwen3-0.6B/config.json "HTTP/1.1 302 Found"
2026-05-25 10:57:39,966 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/README.md "HTTP/1.1 307 Temporary Redirect"
2026-05-25 10:57:39,967 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/Qwen3-0.6B/tokenizer.json "HTTP/1.1 302 Found"
2026-05-25 10:57:39,968 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/Qwen3-0.6B/tokenizer_config.json "HTTP/1.1 302 Found"
2026-05-25 10:57:39,968 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/Qwen3-0.6B%2Fmerges.txt "HTTP/1.1 200 OK"
2026-05-25 10:57:39,970 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/.gitattributes "HTTP/1.1 200 OK"
Downloading (incomplete total...): 0%| | 0.00/2.16k [00:00<?, ?B/s]2026-05-25 10:57:39,977 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/README.md "HTTP/1.1 200 OK"
2026-05-25 10:57:39,987 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/Qwen3-0.6B%2Fmerges.txt "HTTP/1.1 200 OK"
Downloading (incomplete total...): 0%|▏ | 2.16k/1.67M [00:00<01:25, 19.6kB/s2026-05-25 10:57:39,993 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/README.md "HTTP/1.1 200 OK" | 1/21 [00:00<00:02, 9.68it/s]
Downloading (incomplete total...): 1%|█▍ | 15.2k/1.69M [00:00<00:12, 130kB/s]2026-05-25 10:57:40,056 [INFO] HTTP Request: GET https://huggingface.co/api/models/FunAudioLLM/Fun-ASR-Nano-2512/xet-read-token/abb99a38f83a40128cd11bf183315bfa245b69e2 "HTTP/1.1 200 OK"
Downloading (incomplete total...): 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▉| 1.69M/1.69M [00:00<00:00, 130kB/s]2026-05-25 10:57:40,057 [INFO] HTTP Request: GET https://huggingface.co/api/models/FunAudioLLM/Fun-ASR-Nano-2512/xet-read-token/abb99a38f83a40128cd11bf183315bfa245b69e2 "HTTP/1.1 200 OK"
Downloading (incomplete total...): 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▉| 1.69M/1.69M [00:00<00:00, 130kB/s]2026-05-25 10:57:40,059 [INFO] HTTP Request: GET https://huggingface.co/api/models/FunAudioLLM/Fun-ASR-Nano-2512/xet-read-token/abb99a38f83a40128cd11bf183315bfa245b69e2 "HTTP/1.1 200 OK"
Downloading (incomplete total...): 13%|███████████████████▉ | 1.69M/13.1M [00:00<01:27, 130kB/s]2026-05-25 10:57:40,060 [INFO] HTTP Request: GET https://huggingface.co/api/models/FunAudioLLM/Fun-ASR-Nano-2512/xet-read-token/abb99a38f83a40128cd11bf183315bfa245b69e2 "HTTP/1.1 200 OK"
Downloading (incomplete total...): 13%|███████████████████▉ | 1.69M/13.1M [00:00<01:27, 130kB/s]2026-05-25 10:57:40,063 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/README_zh.md "HTTP/1.1 307 Temporary Redirect"
2026-05-25 10:57:40,069 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/Qwen3-0.6B/vocab.json "HTTP/1.1 302 Found"
2026-05-25 10:57:40,070 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/config.yaml "HTTP/1.1 307 Temporary Redirect"
2026-05-25 10:57:40,071 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/README_zh.md "HTTP/1.1 200 OK"
Downloading (incomplete total...): 11%|████████████████▍ | 1.69M/15.9M [00:00<01:49, 130kB/s]2026-05-25 10:57:40,076 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/config.yaml "HTTP/1.1 200 OK"
2026-05-25 10:57:40,079 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/README_zh.md "HTTP/1.1 200 OK"
Downloading (incomplete total...): 11%|████████████████▍ | 1.69M/15.9M [00:00<01:49, 130kB/s]2026-05-25 10:57:40,082 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/config.yaml "HTTP/1.1 200 OK"
Downloading (incomplete total...): 11%|████████████████▌ | 1.70M/15.9M [00:00<01:49, 130kB/s]2026-05-25 10:57:40,084 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/configuration.json "HTTP/1.1 302 Found"
Downloading (incomplete total...): 11%|████████████████▌ | 1.70M/15.9M [00:00<01:49, 130kB/s]2026-05-25 10:57:40,153 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/example/en.mp3 "HTTP/1.1 307 Temporary Redirect"
2026-05-25 10:57:40,158 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fen.mp3 "HTTP/1.1 200 OK"
2026-05-25 10:57:40,159 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/example/ja.mp3 "HTTP/1.1 307 Temporary Redirect"
2026-05-25 10:57:40,163 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fen.mp3 "HTTP/1.1 200 OK"
Downloading (incomplete total...): 11%|████████████████▉ | 1.76M/16.0M [00:00<00:02, 6.98MB/s]2026-05-25 10:57:40,164 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fja.mp3 "HTTP/1.1 200 OK"
2026-05-25 10:57:40,170 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fja.mp3 "HTTP/1.1 200 OK"
Downloading (incomplete total...): 11%|████████████████▉ | 1.76M/16.0M [00:00<00:02, 6.98MB/s]2026-05-25 10:57:40,240 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/example/ko.mp3 "HTTP/1.1 307 Temporary Redirect"
2026-05-25 10:57:40,246 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fko.mp3 "HTTP/1.1 200 OK"
2026-05-25 10:57:40,249 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/example/yue.mp3 "HTTP/1.1 307 Temporary Redirect"
2026-05-25 10:57:40,251 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fko.mp3 "HTTP/1.1 200 OK"
Downloading (incomplete total...): 11%|█████████████████▍ | 1.82M/16.1M [00:00<00:02, 6.98MB/s]2026-05-25 10:57:40,254 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fyue.mp3 "HTTP/1.1 200 OK"
2026-05-25 10:57:40,259 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fyue.mp3 "HTTP/1.1 200 OK"
Downloading (incomplete total...): 11%|█████████████████▋ | 1.85M/16.1M [00:00<00:02, 6.98MB/s]2026-05-25 10:57:40,326 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/example/zh.mp3 "HTTP/1.1 307 Temporary Redirect"
2026-05-25 10:57:40,332 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fzh.mp3 "HTTP/1.1 200 OK"
2026-05-25 10:57:40,337 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/images/compare_en.png "HTTP/1.1 302 Found"
2026-05-25 10:57:40,338 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/example%2Fzh.mp3 "HTTP/1.1 200 OK"
Downloading (incomplete total...): 12%|██████████████████ | 1.92M/16.3M [00:00<00:02, 6.98MB/s]2026-05-25 10:57:40,474 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/images/compare_zh.png "HTTP/1.1 302 Found"
Downloading (incomplete total...): 81%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▌ | 13.4M/16.5M [00:00<00:00, 20.1MB/s]2026-05-25 10:57:40,655 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/images/funasr-v2.png "HTTP/1.1 302 Found" | 2/21 [00:00<00:07, 2.46it/s]
Downloading (incomplete total...): 94%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████ | 16.1M/17.1M [00:00<00:00, 20.1MB/s]2026-05-25 10:57:40,682 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/multilingual.tiktoken "HTTP/1.1 307 Temporary Redirect"
2026-05-25 10:57:40,683 [INFO] HTTP Request: HEAD https://huggingface.co/FunAudioLLM/Fun-ASR-Nano-2512/resolve/abb99a38f83a40128cd11bf183315bfa245b69e2/model.pt "HTTP/1.1 302 Found"
Downloading (incomplete total...): 1%|█▏ | 16.1M/1.99G [00:00<01:37, 20.1MB/s]2026-05-25 10:57:40,687 [INFO] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/multilingual.tiktoken "HTTP/1.1 200 OK"
2026-05-25 10:57:40,692 [INFO] HTTP Request: GET https://huggingface.co/api/resolve-cache/models/FunAudioLLM/Fun-ASR-Nano-2512/abb99a38f83a40128cd11bf183315bfa245b69e2/multilingual.tiktoken "HTTP/1.1 200 OK"
Fetching 21 files: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 21/21 [00:04<00:00, 4.64it/s]
Download complete: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1.99G/1.99G [00:04<00:00, 708MB/s]INFO 05-25 10:57:44 [importing.py:53] Triton module has been replaced with a placeholder.
INFO 05-25 10:57:45 [__init__.py:239] Automatically detected platform cuda.
Download complete: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1.99G/1.99G [00:06<00:00, 297MB/s]
Traceback (most recent call last):
File "/usr/local/src/FunASR/examples/industrial_data_pretraining/fun_asr_nano/serve_realtime_ws.py", line 518, in <module>
asyncio.run(main(args))
File "/usr/lib/python3.12/asyncio/runners.py", line 194, in run
return runner.run(main)
^^^^^^^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/runners.py", line 118, in run
return self._loop.run_until_complete(task)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/base_events.py", line 687, in run_until_complete
return future.result()
^^^^^^^^^^^^^^^
File "/usr/local/src/FunASR/examples/industrial_data_pretraining/fun_asr_nano/serve_realtime_ws.py", line 493, in main
load_models(args)
File "/usr/local/src/FunASR/examples/industrial_data_pretraining/fun_asr_nano/serve_realtime_ws.py", line 409, in load_models
_vllm_engine = AutoModelVLLM(
^^^^^^^^^^^^^^
File "/usr/local/src/FunASR/funasr/auto/auto_model_vllm.py", line 242, in __init__
self._engine = FunASRNanoVLLM(
^^^^^^^^^^^^^^^
File "/usr/local/src/FunASR/funasr/models/fun_asr_nano/inference_vllm.py", line 170, in __init__
from vllm.inputs.data import EmbedsPrompt
ImportError: cannot import name 'EmbedsPrompt' from 'vllm.inputs.data' (/usr/local/src/.venv/lib/python3.12/site-packages/vllm/inputs/data.py). Did you mean: 'TokensPrompt'?
Environment
pip, source, Docker): pipcuda,cpu,mps): cuda报错信息
执行 FunASR/examples/industrial_data_pretraining/fun_asr_nano 目录下的 serve_realtime_ws.py,执行命令如下:
报错信息如下:
看报错似乎是vllm的版本问题,请问这个vllm应该使用哪个版本?或者是不是其它原因导致的?