diff --git a/docs/models/supported_models.md b/docs/models/supported_models.md
index d7b587b6df..584238af9b 100644
--- a/docs/models/supported_models.md
+++ b/docs/models/supported_models.md
@@ -572,6 +572,7 @@ These models primarily accept the [`LLM.generate`](./generative_models.md#llmgen
| `MossAudioModel` | MOSS-Audio | T + A+ | `OpenMOSS-Team/MOSS-Audio-4B-Instruct`, `OpenMOSS-Team/MOSS-Audio-4B-Thinking`, `OpenMOSS-Team/MOSS-Audio-8B-Instruct`, `OpenMOSS-Team/MOSS-Audio-8B-Thinking` | ✅︎ | ✅︎ |
| `MossTranscribeDiarizeForConditionalGeneration` | MOSS-Transcribe-Diarize | T + A | `OpenMOSS-Team/MOSS-Transcribe-Diarize` | | ✅︎ |
| `Moondream3ForCausalLM` | Moondream3 | T + I | `moondream/moondream3-preview` | | ✅︎ |
+| `MuseGlimmerForCausalLM`, `MuseGlimmerForConditionalGeneration` | Muse Glimmer | T + I+ + V+ | `meta-models/Muse-Glimmer-30B` | ✅︎ | ✅︎ |
| `NVLM_D_Model` | NVLM-D 1.0 | T + I+ | `nvidia/NVLM-D-72B`, etc. | | ✅︎ |
| `OpenCUAForConditionalGeneration` | OpenCUA-7B | T + IE+ | `xlangai/OpenCUA-7B` | ✅︎ | ✅︎ |
| `OpenPanguVLForConditionalGeneration` | openpangu-VL | T + IE+ + VE+ | `FreedomIntelligence/openPangu-VL-7B` | ✅︎ | ✅︎ |
@@ -619,6 +620,13 @@ Some models are supported only via the [Transformers modeling backend](#transfor
* Only specific variants of the model support this modality (see notes below).
Q `Qwen*-VL` officially uses `qwen_vl_utils` for image preprocessing, while vLLM uses `transformers`' `video_processing_qwen*`, which leads to slightly different results compared to the official Hugging Face repository examples.
+!!! note
+ For `Dots3NoteForCausalLM`, the vision and audio towers are only loaded when the
+ corresponding modality is enabled via `--limit-mm-per-prompt`. Video inputs are
+ decoded into frames and audio, so they require both towers. The checkpoint also
+ ships one MTP layer, enabled with
+ `--speculative-config '{"method":"mtp","num_speculative_tokens":1}'`.
+
!!! note
`Gemma3nForConditionalGeneration` is only supported on V1 due to shared KV caching and it depends on `timm>=1.0.17` to make use of its
MobileNet-v5 vision backbone.
@@ -656,6 +664,13 @@ Some models are supported only via the [Transformers modeling backend](#transfor
coordinate decoding and are not exposed by this vLLM implementation.
See [Moondream3 prompt recipes](../features/multimodal_inputs.md#moondream3-prompt-recipes).
+!!! note
+ Both Muse Glimmer architecture names map to the same vLLM implementation:
+ checkpoints with a vision config accept image and video inputs, while
+ vision-less checkpoints run as a text-only model. Speculative decoding uses the
+ `meta-models/Muse-Glimmer-30B-assistant` checkpoint, which vLLM serves through
+ its [DFlash](../features/speculative_decoding/README.md) path.
+
!!! note
The official `openbmb/MiniCPM-V-2` doesn't work yet, so we need to use a fork (`HwwwH/MiniCPM-V-2`) for now.
For more details, please see:
diff --git a/tests/models/registry.py b/tests/models/registry.py
index 62cb289e4f..16604ec7ff 100644
--- a/tests/models/registry.py
+++ b/tests/models/registry.py
@@ -847,8 +847,7 @@ _MULTIMODAL_EXAMPLE_MODELS = {
"deepseek-ai/DeepSeek-OCR-2",
),
"Dots3NoteForCausalLM": _HfExamplesInfo(
- "rednote-hilab/dots3.note",
- trust_remote_code=True,
+ "dots-studio/dots3-note-prev",
is_available_online=False,
),
"UnlimitedOCRForCausalLM": _HfExamplesInfo(
@@ -1681,9 +1680,8 @@ _SPECULATIVE_DECODING_EXAMPLE_MODELS = {
is_available_online=False,
),
"Dots3NoteMTPModel": _HfExamplesInfo(
- "rednote-hilab/dots3.note",
- speculative_model="rednote-hilab/dots3.note",
- trust_remote_code=True,
+ "dots-studio/dots3-note-prev",
+ speculative_model="dots-studio/dots3-note-prev",
is_available_online=False,
),
"Gemma4MTPModel": _HfExamplesInfo(