A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 3, 22:45 UTC
Document MolmoWeb hf_overrides (#42163)
Signed-off-by: Haoqi Wang <78337154+hqhq1025@users.noreply.github.com>
Haoqing Wang committed
May 11, 2026 at 14:58 UTC
5cba6839e6cb8ede0c81947557b76617468d0d61
1 file changed
+6
-1
docs/models/supported_models.md
+6
-1
@@ -598,7 +598,7 @@ These models primarily accept the [`LLM.generate`](./generative_models.md#llmgen
598
| `MiniMaxVL01ForConditionalGeneration` | MiniMax-VL | T + I<sup>E+</sup> | `MiniMaxAI/MiniMax-VL-01`, etc. | | ✅︎ |
599
| `Mistral3ForConditionalGeneration` | Mistral3 (HF Transformers) | T + I<sup>+</sup> | `mistralai/Mistral-Small-3.1-24B-Instruct-2503`, etc. | ✅︎ | ✅︎ |
600
| `MolmoForCausalLM` | Molmo | T + I<sup>+</sup> | `allenai/Molmo-7B-D-0924`, `allenai/Molmo-7B-O-0924`, etc. | ✅︎ | ✅︎ |
601
-| `Molmo2ForConditionalGeneration` | Molmo2 | T + I<sup>+</sup> / V | `allenai/Molmo2-4B`, `allenai/Molmo2-8B`, `allenai/Molmo2-O-7B` | ✅︎ | ✅︎ |
601
+| `Molmo2ForConditionalGeneration` | Molmo2 | T + I<sup>+</sup> / V | `allenai/Molmo2-4B`, `allenai/Molmo2-8B`, `allenai/Molmo2-O-7B`, `allenai/MolmoWeb-4B`<sup>^</sup>, `allenai/MolmoWeb-8B`<sup>^</sup> | ✅︎ | ✅︎ |
602
| `Moondream3ForCausalLM` | Moondream3 | T + I | `moondream/moondream3-preview` | | ✅︎ |
603
| `MusicFlamingoForConditionalGeneration` | MusicFlamingo | T + A | `nvidia/music-flamingo-2601-hf`, `nvidia/music-flamingo-think-2601-hf` | ✅︎ | ✅︎ |
604
| `NVLM_D_Model` | NVLM-D 1.0 | T + I<sup>+</sup> | `nvidia/NVLM-D-72B`, etc. | | ✅︎ |
@@ -666,6 +666,11 @@ Some models are supported only via the [Transformers modeling backend](#transfor
666
!!! note
667
For `InternVLChatModel`, only InternVL2.5 with Qwen2.5 text backbone (`OpenGVLab/InternVL2.5-1B` etc.), InternVL3 and InternVL3.5 have video inputs support currently.
668
669
+!!! note
670
+ To use `allenai/MolmoWeb-4B` or `allenai/MolmoWeb-8B`, serve the checkpoint
671
+ with the Molmo2 architecture and disable multimodal-prefix attention:
672
+ `--hf-overrides '{"architectures": ["Molmo2ForConditionalGeneration"], "is_mm_prefix_lm": false}'`.
673
+
674
!!! note
675
`Moondream3ForCausalLM` uses task-specific prompt templates for `query`
676
and `caption`. The native `detect` and `point` skills require custom