MolmoAct2 is an open vision-language-action model for robot control. It builds on Molmo2-ER, an embodied-reasoning VLM backbone, and connects the autoregressive VLM to a flow-matching continuous action expert through per-layer KV (key-value) conditioning. This checkpoint is the post-trained, multi-embodiment MolmoAct2 model. It is intended as a foundation checkpoint for further robot fine-tuning rather than as a ready-to-run policy for a single deployment setting. Use this checkpoint for further fine-tuning on a target robot embodiment or benchmark. It contains the VLM and continuous action expert weights, plus normalization metadata for the post-training mixture in normstats.json. This…
MolmoAct2-Think extends MolmoAct2 with depth-token reasoning. Before producing an action, the model can predict a compact 10 x 10 discrete depth representation and condition the action expert on the resulting depth-aware VLM cache.
Parameters5.4B
Context16,384
Weights21.8 GB
License—
AccessOpen weights
Monthly Downloads18.9k
Runs On
What it takes to serve MolmoAct2-Think (5.4B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 10.9 GB | 13.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 5.4 GB | 6.5 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 2.7 GB | 3.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.
Model Card
MolmoAct2-Think extends MolmoAct2 with depth-token reasoning. Before producing an action, the model can predict a compact 10 x 10 discrete depth representation and condition the action expert on the resulting depth-aware VLM cache. This checkpoint is the post-trained, multi-embodiment depth-reasoning model. It is intended as a foundation checkpoint for further robot fine-tuning rather than as a ready-to-run policy for a single deployment setting. Use this checkpoint for further fine-tuning when the downstream policy should use depth reasoning. It contains the VLM, action expert, and depth-token weights, plus normalization metadata for the post-training mixture in normstats.json. This model…
Excerpt from the card by Ai2.
Configuration
- Architecture
- MolmoAct2ForConditionalGeneration
- Context length (tokens)
- 16,384
- Layers
- 36
- Hidden size
- 2,560
- Feed-forward size
- 9,728
- Attention heads
- 32
- Key/value heads
- 8
- Head dimension
- 128
- Vocabulary size
- 155,648
- RoPE base
- 5e+06
- Model type
- molmoact2
Identity and Version
- Repository
- allenai/MolmoAct2-Think
- Publisher
- Ai2
- Task
- Robotics
- Modality
- Control
- Library
- transformers
- Parameters
- 5.4B parameters
- Languages
- Not stated by the source
- Revision
- 94822d94d5fb033714a44838638de0c137872b99
- First published
- 2026-05-04
- Last updated
- 2026-05-23
Files and Weights
22 files, 21.8 GB in total. The weights are 5 files totalling 21.8 GB in safetensors.
Weights5 files · 21.8 GB
Configuration11 files · 496.0 KB
Tokenizer2 files · 12.2 MB
Documentation1 file · 3.5 KB
Other2 files · 12.8 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00005.safetensors | Weights | 4.9 GB | 08663a88bc5c |
| model-00002-of-00005.safetensors | Weights | 4.8 GB | 9282d1390ca3 |
| model-00003-of-00005.safetensors | Weights | 4.8 GB | 3f35fdd96126 |
| model-00004-of-00005.safetensors | Weights | 4.9 GB | c58018140b07 |
| model-00005-of-00005.safetensors | Weights | 2.3 GB | 2b36561311b5 |
| config.json | Configuration | 4.3 KB | — |
| configuration_molmoact2.py | Configuration | 20.7 KB | — |
| generation_config.json | Configuration | 116 B | — |
| image_processing_molmoact2.py | Configuration | 22.4 KB | — |
| inference.py | Configuration | 25.9 KB | — |
| model.safetensors.index.json | Configuration | 131.5 KB | — |
| modeling_molmoact2.py | Configuration | 191.0 KB | — |
| norm_stats.json | Configuration | 43.8 KB | — |
| processing_molmoact2.py | Configuration | 17.3 KB | — |
| processor_config.json | Configuration | 1.9 KB | — |
| video_processing_molmoact2.py | Configuration | 37.2 KB | — |
| README.md | Documentation | 3.5 KB | — |
| assets/MolmoAct2-Think.svg | Other | 9.6 KB | — |
| chat_template.jinja | Other | 3.2 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 12.2 MB | 8b6aeec78de2 |
| tokenizer_config.json | Tokenizer | 782 B | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 21.8 GB
Download from Ai2
Released by Ai2 through its official repository on Hugging Face.
Built From
- Described by arXiv:2605.02881
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 21.8 GB |
| 16-bit | 10.9 GB |
| 8-bit | 5.4 GB |
| 4-bit | 2.7 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About MolmoAct2-Think
How much GPU memory does MolmoAct2-Think need?
About 13.1 GB at 16-bit and 3.3 GB at 4-bit: the weights (5.4B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run MolmoAct2-Think on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
What is MolmoAct2-Think's context length?
16,384 tokens, from the maximum position embeddings in its published configuration.
Similar Models
MolmoAct2 is an open vision-language-action model for robot control. It builds on Molmo2-ER and attaches a flow-matching continuous action expert that conditions on the VLM key-value cache through a per-layer connection. This checkpoint is fine-tuned on the SO-100/101 mixture with absolute joint-pose control and annotated language instructions. It is intended for both further fine-tuning and SO-100/101 policy inference. Use this checkpoint for SO-100/101 inference or for further fine-tuning. Dataset normalization metadata is stored in normstats.json. pass normtag="so100so101molmoact2" at inference time. Continuous action prediction is the intended and recommended inference mode. Discrete…
MolmoAct2 is an open vision-language-action model for robot control. It builds on Molmo2-ER and attaches a flow-matching continuous action expert that conditions on the VLM key-value cache through a per-layer connection. This checkpoint is fine-tuned on the full LIBERO training mixture, combining Spatial, Object, Goal, and Long suites. It is intended for both further fine-tuning and LIBERO policy inference. Use this checkpoint for LIBERO inference or for further fine-tuning. Dataset normalization metadata is stored in normstats.json. pass normtag="libero" at inference time. Continuous action prediction is the intended and recommended inference mode. Discrete action prediction is exposed for…
MolmoAct2 is an open vision-language-action model for robot control. It builds on Molmo2-ER and attaches a flow-matching continuous action expert that conditions on the VLM key-value cache through a per-layer connection. This checkpoint is fine-tuned on the bimanual YAM mixture with absolute joint-pose control and annotated language instructions. It is intended for both further fine-tuning and bimanual YAM policy inference. Use this checkpoint for bimanual YAM inference or for further fine-tuning. Dataset normalization metadata is stored in normstats.json. pass normtag="yamdualmolmoact2" at inference time. Continuous action prediction is the intended and recommended inference mode. Discrete…
This policy has been trained and pushed to the Hub using LeRobot. See the full documentation at LeRobot Docs. For a complete walkthrough, see the training guide. Below is the short version on how to train and run inference/eval: Writes checkpoints to outputs/train/ /checkpoints/. Prefix the dataset repo with eval\ and supply --policy.path pointing to a local or hub checkpoint.
This policy has been trained and pushed to the Hub using LeRobot. See the full documentation at LeRobot Docs. For a complete walkthrough, see the training guide. Below is the short version on how to train and run inference/eval: Writes checkpoints to outputs/train/ /checkpoints/. Prefix the dataset repo with eval\ and supply --policy.path pointing to a local or hub checkpoint.