Microsoft Research
CogACT
Commercial use OK 7/8 verified
1 of 8 required fields are unverified. They are shown as badges, not filled with defaults. Sources for verified fields are listed below.
Facts
- Released
- 2024-12 1 Source
- Parameters
- Unverified 0 Source
- Architecture
- VLM + action expert 0 Source
- License
- MIT (code, weights and data all MIT, per README) 2 Source
- Commercial
- Commercial use OK 0 Source
- Weights
- huggingface.co ↗ 1 Source
- Code
- github.com ↗ 1 Source
- Paper
- arxiv.org ↗ 0 Source
- Input modalities
- rgb · language 0 Source
- Action space
- 7-DoF end-effector pose delta + gripper (diffusion, 16-step action chunks) 0 Source
- Cross-embodiment
- Open X-Embodiment pretraining 0 Source
- Pretraining data (hours)
- Unverified 0 Source
- Target forms
- Single arm 0 Source
Architecture · VLM backbone (prism-dinosiglip-224px) + DiT diffusion action module (three sizes DiT-S/B/L); outputs 16 steps of 7-DoF actions at once
Sources · 2 sources
license,release_date,arch_notes,code_url https://github.com/microsoft/CogACT Official page · Fetched 2026-09-04 · No snapshot stored yet README：released 2024-12-01；All the code, model weights, and data are licensed under MIT license；CogACT-Small/Base/Large，DiT-S/B/L
weights_url,license https://huggingface.co/api/models/CogACT/CogACT-Base Model hub page · Fetched 2026-09-04 · No snapshot stored yet Model card license: mit; createdAt 2024-11-29
Activity signals
Public counters fetched daily from GitHub, Hugging Face and ModelScope. Numbers only; no judgement. · fetched 2026-10-31
GitHub stars
434
+2 / 30 d
Commits · 90 d
0
last push 367 d ago
HF downloads · 30 d
1,711
all-time 80,220 · 19 likes
Weights last modified
2024-12-04
697 d ago
ModelScopeno same-name mirror found (searched by repo name)
Hugging FaceCogACT/CogACT-Base
stars · last 30 d
Public benchmark scores
As printed in the source; protocols differ between papers. Full table with notes on the benchmarks page.
| Benchmark | Subset | Score | Source |
|---|---|---|---|
| Real-robot evaluation (self-reported) | franka_4_tasks_avg | 61.4 | paper · Table 6 (real robot, Franka), row 'CogACT', Average |
| Real-robot evaluation (self-reported) | realman_arm_pick_stack_place_overall_avg | 71.2 | paper · Table 3 (real robot, Realman), row 'CogACT', Overall Avg |
| SimplerEnv · Google RobotCogACT-Base (DiT-Base action module) | variant_aggregation | 61.3 | paper · Table 1 (Variant Aggregation), row 'CogACT', Average |
| SimplerEnv · Google RobotCogACT-Large (DiT-Large, 308M) | variant_aggregation | 59.3 | paper · Table 7, row 'DiT-Large', Google VA |
| SimplerEnv · Google RobotCogACT-Base (DiT-Base action module) | visual_matching | 74.8 | paper · Table 1 (Visual Matching), row 'CogACT', Average |
| SimplerEnv · Google RobotCogACT-Large (DiT-Large, 308M) | visual_matching | 76.7 | paper · Table 7 (action model scaling), row 'DiT-Large', Google VM |
| SimplerEnv · Google RobotCogACT-Small (DiT-Small, 13M) | visual_matching | 73.3 | paper · Table 7, row 'DiT-Small', Google VM |
| SimplerEnv · WidowX (Bridge)CogACT-Base (DiT-Base action module) | visual_matching | 51.3 | paper · Table 2 (WidowX robot), row 'CogACT', Average |
| SimplerEnv · WidowX (Bridge)CogACT-Large (DiT-Large, 308M) | visual_matching | 58.3 | paper · Table 7, row 'DiT-Large', WidowX VM |
Deployment recipe
No recipe yet. We publish only steps that actually ran on a rented GPU with a measured latency; this model has not been through that yet.
Record history
2026-09-04
Sources fetched
license,release_date,arch_notes,code_url · weights_url,license
Latency & cost
cost = p50 ms ÷ 1000 × rental ÷ 3600 × 1000
| Hardware | p50 ms | p95 ms | Peak VRAM | Cost per 1k inferences ¥ | samples |
|---|---|---|---|---|---|
NVIDIA RTX 4090 ¥1.68/h · GongJi Compute (2026) | — | — | — | — no latency data | No data |
NVIDIA RTX 5090 ¥2.4/h · Runyun 2.4 / GongJi 2.5 / Paratera Cloud 2.98 (2026) | — | — | — | — no latency data | No data |
NVIDIA A100 80GB no rental price | — | — | — | — no latency data | No data |
NVIDIA Jetson AGX Thor no rental price | — | — | — | — no latency data | No data |
NVIDIA H100 80GB no rental price | — | — | — | — no latency data | No data |
NVIDIA A800 80GB no rental price | — | — | — | — no latency data | No data |
No latency measurements for this model yet · Data comes from literature, maintainer runs, or crowd CLI uploads. None exist so far.
Compatibility
Colored only where we hold evidence. An amber dot means the vendor declares this form as a target; that is a claim, not a verified compatibility.
| Unitree G1 | Unitree H1 | AgiBot Genie G1 | AgileX Piper arm | SO-101 | Franka FR3 / Panda | Fourier GR-2 | WidowX 250 S | Google Robot（Everyday Robots） | ALOHA 2 / Cobot Magic (dual-arm) | Galaxea R1 Pro | Franka Emika Panda | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CogACT |
Legend Officially supportedCommunity reproducedFine-tunable in theoryUnsupportedUnknown Vendor-declared target form 3 / 12 cells with a definite status
Reproduction records
Did this model run on your robot? Report it with a link (log, video or repo). Records are reviewed by the maintainers; a verified “works” or “works after fine-tuning” upgrades the matrix cell to “community verified” with your link as evidence. Anonymous is fine; nothing but the fields below is stored.
No verified record yet for this model.
Clear direction, grounded judgement.