[Docs][Models] Use the official FunASR Nano vLLM checkpoint (#54944)

Signed-off-by: LauraGPT <18321252+LauraGPT@users.noreply.github.com> Co-authored-by: LauraGPT <18321252+LauraGPT@users.noreply.github.com> Co-authored-by: OpenAI Codex <codex@openai.com>

zhifu gao committed Sep 5, 2026 at 15:29 UTC e473e9036f979d546830aece9855027049faf0ba
2 files changed +6 -5
docs/models/supported_models.md
+1 -1
@@ -711,7 +711,7 @@ Speech2Text models trained specifically for Automatic Speech Recognition.
711 | ------------ | ------ | ----------------- | -------------------- | ------------------------- |
712 | `CohereAsrForConditionalGeneration` | Cohere-Transcribe | `CohereLabs/cohere-transcribe-03-2026` | | |
713 | `FireRedASR2ForConditionalGeneration` | FireRedASR2 | `allendou/FireRedASR2-LLM-vllm`, etc. | | |
714 -| `FunASRForConditionalGeneration` | FunASR | `allendou/Fun-ASR-Nano-2512-vllm`, etc. | | |
714 +| `FunASRForConditionalGeneration` | FunASR | `FunAudioLLM/Fun-ASR-Nano-2512-vllm`, etc. | | |
715 | `Gemma3nForConditionalGeneration` | Gemma3n | `google/gemma-3n-E2B-it`, `google/gemma-3n-E4B-it`, etc. | | |
716 | `GlmAsrForConditionalGeneration` | GLM-ASR | `zai-org/GLM-ASR-Nano-2512` | ✅︎ | ✅︎ |
717 | `GraniteSpeechForConditionalGeneration` | Granite Speech | `ibm-granite/granite-4.0-1b-speech`, `ibm-granite/granite-speech-3.3-2b`, etc. | ✅︎ | ✅︎ |
tests/models/registry.py
+5 -4
@@ -876,12 +876,13 @@ _MULTIMODAL_EXAMPLE_MODELS = {
876 },
877 ),
878 "FunASRForConditionalGeneration": _HfExamplesInfo(
879 - "allendou/Fun-ASR-Nano-2512-vllm",
879 + "FunAudioLLM/Fun-ASR-Nano-2512-vllm",
880 trust_remote_code=True,
881 - max_transformers_version="5.1",
881 + max_transformers_version="5.15",
882 transformers_version_reason={
883 - "vllm": "Incompatible with transformers v5.2+ "
884 - "(dict object has no attribute '__name__').",
883 + "vllm": "The official Fun-ASR-Nano vLLM package is validated "
884 + "through transformers v5.15.0; newer versions require separate "
885 + "validation.",
886 },
887 ),
888 "FunAudioChatForConditionalGeneration": _HfExamplesInfo(