Cosmos-Reason2-32B
nvidia/Cosmos-Reason2-32B
Tracked by OpenModelStats since September 4, 2026View on Hugging Face ↗
- Downloads 30DRolling recent-window download count as reported by the source platform.
- 31.4K
- Total downloadsCumulative all-time download counter reported by the source platform.
- 51.6K
- Gained 7DIncrease in cumulative total downloads between OpenModelStats' own observations seven days apart.
- +7,626(increase)
- 17.9%(up) vs baseline
- Gained 30DIncrease in cumulative total downloads between OpenModelStats' own observations roughly thirty days apart.
- —
- LikesCommunity likes reported by the source platform.
- 14
- +0 in 7D
- Rank · tracked modelsPosition among all eligible models tracked by OpenModelStats, ordered by recent downloads. #602 among tracked Image Text to Text models.
- #4,720
- 9(up 9 places)this week
- Parameters
- 33.4B
- Last source update
- Apr 30, 2026
- MomentumOpenModelStats' composite rising-model signal (0–100): a percentile blend of absolute 7-day download gains, capped percentage growth, and likes gained. See Methodology.
- 71.2
- Cosmos-Reason2-32B gained 7,626 downloads during the last seven tracked days.
- It moved from #603 to #602 among tracked Image Text to Text models this week.
Downloads over time
Rolling recent-window downloads as reportedLikes over time
Rank among tracked models
Lower is betterOverview
Cosmos-Reason2-32B is a image text to text model published by nvidia. OpenModelStats has tracked the model since Sep 4, 2026, most recently observing it Oct 4, 2026.
- Published
- Apr 29, 2026
- Last updated
- Apr 30, 2026
- Library
- cosmos
- License
- other
- Task
- Image Text to Text
- Parameters
- 33,357,390,064
- Spaces
- —
- Derivative models
- 1
- Velocity 7D/day
- 1,089
cosmossafetensorsqwen3_vlnvidiaconversationalimage-text-to-textarxiv:2503.06800arxiv:2406.10721arxiv:2603.18178arxiv:2312.14115base_model:Qwen/Qwen3-VL-32B-Instructbase_model:finetune:Qwen/Qwen3-VL-32B-Instruct
Publisher
nvidiaView publisher statistics →Derived from Qwen/Qwen3-VL-32B-Instruct (reported by the source; parent not yet tracked).
Model family
Based on base-model relationships reported by the source's metadata.
Derivative models
Similar models
| Model | DL 30D |
|---|---|
| Qwen3-VL-32B-InstructQwen | 335.3K |
| Qwen3-VL-32B-Instruct-AWQQuantTrio | 151.2K |
| Qwen3-VL-32B-Instruct-heretic-v2coder3101 | 58.8K |
| Qwen3-VL-32B-ThinkingQwen | 16.9K |
| Qwen3-VL-32B-Instruct-unsloth-bnb-4bitunsloth | 1,677 |
| Qwen3-VL-32B-Instruct-bnb-4bitunsloth | 1,258 |