MolmoBot
Robotics modelRobot manipulation policies and evaluation code paired with simulated training data and the MolmoSpaces generation engine. [1] [3]
Setup & usage →CheckpointsModel hub / homepageSource codeGetting started
- For
- Robot learning researchers comparing simulation-trained manipulation policies. [1]
- Parameters
- ~4B VLM backbone (Molmo2-4B) + per-layer action expert (18.62 GB PyTorch checkpoint for MolmoBot-DROID [6] [8]
- Embodiments
- DROID (Franka Emika Panda via Polymetis), Rainbow Robotics RBY1 (dedicated RBY1 checkpoints), MolmoSpaces / MuJoCo… [1] [3] [9]
- Source rights
- Named licenses are documented by part. Review licenses
Source researched · Not tested by us
Sources checked · Research details
Overview
MolmoBot (allenai/MolmoBot) is a family of manipulation policies from the Allen Institute for AI trained on procedurally generated simulation episodes (allenai/MolmoBot-Data via MolmoSpaces) without real-robot training data for its primary DROID checkpoint. At the pinned Hugging Face revision (cbe6ec358958), allenai/MolmoBot-DROID (18.62 GB PyTorch model.pt on disk) builds on allenai/Molmo2-4B (Qwen3-4B-Instruct + SigLIP 378px) with a per-layer cross-attention flow-matching action expert (action_dim: 8, action_horizon: 16, n_action_steps: 8, flow_matching_num_steps: 10, quantile normalization). [1] [6] [8]
Model card summary
- Architecture
Flow-matching VLA initialized from Molmo2-4B (Qwen3-4B-Instruct LLM + SigLIP 378px vision encoder) with a per-layer cross-attention action expert (10 flow-matching steps); companion variants include MolmoBot-Pi0 (PaliGemma-3B) and MolmoBot-SPOC.
- Base model
allenai/Molmo2-4B (MolmoBot-DROID / RBY1); google/paligemma-3b-mix-224 (MolmoBot-Pi0-DROID)
- Parameter count
~4B VLM backbone (Molmo2-4B) + per-layer action expert (18.62 GB PyTorch checkpoint for MolmoBot-DROID; exact safetensors count not published)
Inferred from Molmo2-4B base model and 18.62 GB model.pt; binaries not downloaded.
- Input modalities
- rgbproprioceptionlanguage
Wrist camera + 1 exo RGB camera, robot joint state (observation.state), and natural-language task instruction.
- Output modalities
- joint-positionsgripper-command
Joint-position + gripper actions (--action-type joint_pos, action_dim: 8 for DROID Franka).
- Action space
Continuous flow-matching action chunks (action_dim: 8, action_horizon: 16, n_action_steps: 8, flow_matching_num_steps: 10, quantile normalization) for DROID joint-position and gripper control.
- Trained / supported embodiments
- DROID (Franka Emika Panda via Polymetis)
- Rainbow Robotics RBY1 (dedicated RBY1 checkpoints)
- MolmoSpaces / MuJoCo simulation
- Training datasets
- allenai/MolmoBot-Data (procedurally generated simulation episodes via MolmoSpaces; zero real-robot data for MolmoBot-DROID)
Checkpoints
| Checkpoint | Format | Size | Access | Pinned revision |
|---|---|---|---|---|
| allenai/MolmoBot-DROID | pytorch | 18.62 GB | open | cbe6ec358958 |
| allenai/MolmoBot-RBY1Multitask | pytorch | 18.62 GB | open | 0786434b1014 |
| allenai/MolmoBot-Pi0-DROID | safetensors | 6.53 GB | open | 48c8e2fc1868 |
| allenai/MolmoBot-SPOC-DROID | safetensors | 0.41 GB | open | 69b48b801478 |
Four representative ungated checkpoints verified via Hugging Face API; note variant licenses differ (Apache-2.0 vs Gemma vs Other).
Is this for you?
- Use cases
- Simulation policy evaluation
- DROID manipulation policy research
- Known limitations
- Paper performance is author-reported and not reproduced here.
- DROID evaluation needs matching hardware, cameras, Polymetis servers and checkpoint.
- Dataset scenes carry object-specific reuse considerations.
- Safety and limitations
- DROID deployment requires matching Franka/Polymetis robot and gripper servers, wrist + exo cameras, and --action-type joint_pos.
- Companion checkpoints carry different licenses (Apache-2.0 for MolmoBot-DROID/RBY1, Gemma for MolmoBot-Pi0-DROID, Other for MolmoBot-SPOC-DROID).
- MolmoBot-Data carries an ODC-BY dataset card, while individual simulated 3D scene objects have asset-specific licenses; commercial-use selection should consult the provided non-NC episode filter and applicable object terms.
Compute & fine-tuning
- Inference compute
- Documented hardware: CUDA GPU (depends on variant; 18.62 GB model.pt for MolmoBot-DROID)
- Edge / consumer capable: No
Exact minimum inference VRAM and latency_ms not stated upstream.
- Fine-tuning
- Supported: Yes
- Methods: Full training / fine-tuning from Molmo2-4B via torchrun (8-64 H100s in upstream recipe; micro-batching supported)
- Recipe: Upstream fine-tuning documentation
- Host computer
GPU resources depend on policy variant and task; no independently measured minimum.
- Operating system
- Linux
uv / MuJoCo EGL / Polymetis Linux workflow documented.
[3] · Editorial assessment
- Dependencies
- MolmoSpaces for data generation
- DROID/Polymetis for documented physical evaluation
Setup & usage
Open the upstream getting-started guide [1] · Model card / documentation [1]
Upstream-reported benchmarks
Upstream-reported only: Open Source Robotics has not run training, inference, or hardware evaluations for these benchmarks.
| Benchmark | Upstream-reported score | Scope & caveats |
|---|---|---|
| Zero-shot sim-to-real manipulation on DROID and RBY1 (arXiv:2603.16861) | See arXiv:2603.16861 (upstream reports zero-shot sim-to-real transfer trained purely on MolmoBot-Data simulation episodes) | Author-reported in arXiv:2603.16861; not independently verified by OSR. |
Project notes
Checkpoints, workflows, and companion variants
Upstream publishes multiple ungated checkpoints across Franka DROID and Rainbow Robotics RBY1 embodiments, including allenai/MolmoBot-DROID (18.62 GB), allenai/MolmoBot-RBY1Multitask (18.62 GB), allenai/MolmoBot-Pi0-DROID (6.53 GB safetensors, PaliGemma-3B base), and allenai/MolmoBot-SPOC-DROID (0.41 GB safetensors). For software inspection, MolmoBot/README.md documents uv sync --extra eval, a simulation demo notebook, and MuJoCo/MolmoSpaces evaluation. Physical DROID execution requires Polymetis robot and gripper servers, one wrist and one external RGB camera, a policy server (serve_molmo.py --hf-repo allenai/MolmoBot-DROID --action-type joint_pos), and the robot evaluation client. Minimum inference VRAM, inference latency, and deployment control frequency are not established by this review, and checkpoint download sizes are disk sizes rather than GPU-memory requirements. [1] [3] [8] [9]
Training compute and upstream-reported benchmarks
Upstream's replication recipe trains from Molmo2-4B using torchrun on 8–64 H100 GPUs (noting micro-batching support for smaller GPUs). Zero-shot sim-to-real transfer results on DROID and RBY1 in arXiv:2603.16861 are author-reported; Open Source Robotics has not downloaded checkpoint binaries or data shards, and has not run training, fine-tuning, inference, simulation, or robot evaluations. [1] [3] [6]
What we checked
Source research only. No commercial relationship or hands-on testing. No upstream media copied to the public site.
Not yet established
- Documented closed-loop control frequency (Hz)
Updates
- Upstream last updated
- Apr 27, 2026 [1]
- Latest release
- Not established
- Profile updated
- Nov 21, 2026
- Sources last checked
- Nov 21, 2026
- Next review
- Feb 19, 2027
- Reviewed revision
- Commit snapshot 33c0ca77bf6062a23d60ffd4a6859334c4a46d30; no matching GitHub release found · combined hardware and software repository
Separate license matrix
Inspectable robot-specific policy implementation with simulation and DROID workflows; data-generation engine, weights and dataset retain separate scope and conditions.
| Layer | Availability / license |
|---|---|
| Code | available · Apache-2.0 Artifact and declaration inspected at the cited scope. External assets/dependencies and actual operation need separate verification. [3] [2] |
| Model weights | available · Apache-2.0 (MolmoBot-DROID / RBY1); Gemma (MolmoBot-Pi0-DROID); Other (MolmoBot-SPOC-DROID) Primary Molmo2-4B-based checkpoints declare Apache-2.0, while companion Pi0 and SPOC variants carry Gemma and Other license tags on Hugging Face Hub. [6] [8] [9] |
| Training datasets | available · ODC-BY card declaration; object-level restrictions require filtering Artifact and declaration inspected at the cited scope. External assets/dependencies and actual operation need separate verification. [7] |
| Documentation | available · Apache-2.0 Artifact and declaration inspected at the cited scope. External assets/dependencies and actual operation need separate verification. [1] [2] |
Separate code, variant weight, and dataset rights
The allenai/MolmoBot code repository is Apache-2.0, and the primary Molmo2-4B checkpoints (MolmoBot-DROID, MolmoBot-RBY1Multitask) declare Apache-2.0. Companion checkpoints carry different Hugging Face license tags (gemma for MolmoBot-Pi0-DROID, other for MolmoBot-SPOC-DROID). allenai/MolmoBot-Data declares ODC-BY at the dataset card level, while individual simulated 3D scene objects carry asset-specific licenses; users selecting episodes for commercial use should consult the provided non-NC episode filter and applicable object licenses rather than treating the dataset label alone as blanket clearance. [2] [6] [7] [9]
References
Each source links to the exact revision we read. “Used for” shows where this page relies on it.
- S001 MolmoBot overview
Revision, date & where cited
allenai/MolmoBot contributors · github.com · revision
33c0ca7· README.md at · Accessed Nov 7, 2026
Used for: Overview, Model card summary, Is this for you?, Compute & fine-tuning and 6 more - S002 Apache-2.0 source license
Revision, date & where cited
allenai/MolmoBot contributors · github.com · revision
33c0ca7· LICENSE at · Accessed Nov 7, 2026
Used for: Separate license matrix - S003 Policy installation and evaluation
Revision, date & where cited
allenai/MolmoBot contributors · github.com · revision
33c0ca7· MolmoBot/README.md at · Accessed Nov 7, 2026
Used for: Model card summary, Is this for you?, Compute & fine-tuning, Setup & usage and 3 more - S004 Pi0 variant documentation
Revision, date & where cited
allenai/MolmoBot contributors · github.com · revision
33c0ca7· MolmoBot-Pi0/README.md at · Accessed Nov 7, 2026 - S005 SPOC variant documentation
Revision, date & where cited
allenai/MolmoBot contributors · github.com · revision
33c0ca7· MolmoBot-SPOC/README.md at · Accessed Nov 7, 2026 - S006 DROID checkpoint card
Revision, date & where cited
Allen Institute for AI · huggingface.co · Model card at · Accessed Nov 7, 2026
Used for: Overview, Model card summary, Is this for you?, Upstream-reported benchmarks and 2 more - S007 MolmoBot-Data card
Revision, date & where cited
Allen Institute for AI · huggingface.co · Dataset card at · Accessed Nov 7, 2026
Used for: Model card summary, Is this for you?, Separate license matrix - S008 Hugging Face API metadata and config.yaml: allenai/MolmoBot-DROID
18.62 GB PyTorch model.pt, ungated, license: apache-2.0, base_model: allenai/Molmo2-4B (Qwen3-4B-Instruct + SigLIP 378px), action_dim: 8, action_horizon: 16, n_action_steps: 8, flow_matching_num_steps: 10, quantile normalization, lastModified 2026-03-21.Revision, date & where cited
Hugging Face (allenai) · huggingface.co · HF API & config.yaml at revision · Accessed Nov 21, 2026
Used for: Overview, Model card summary, Checkpoints, Compute & fine-tuning and 2 more - S009 Hugging Face API metadata: companion MolmoBot checkpoints
MolmoBot-RBY1Multitask: 18.62 GB .pt, apache-2.0, base allenai/Molmo2-4B; MolmoBot-Pi0-DROID: 6.53 GB safetensors, gemma, base google/paligemma-3b-mix-224; MolmoBot-SPOC-DROID: 0.41 GB safetensors, other.Revision, date & where cited
Hugging Face (allenai) · huggingface.co · HF API metadata for MolmoBot-RBY1Multitask (0786434b), MolmoBot-Pi0-DROID (48c8e2fc), MolmoBot-SPOC-DROID (69b48b80) · Accessed Nov 21, 2026
Used for: Model card summary, Checkpoints, Is this for you?, Checkpoints, workflows, and companion variants and 1 more
Correction guidance · Discussion is not enabled · Profile format 0.8.0-draft.1
Change history
2026-11-21 — Migrated 0.7.0-draft.5 -> 0.8.0-draft.1 (kind=model); added verified config.yaml (S008) and companion checkpoint metadata (S009) while preserving historical source dates.