GameScaling-2B-S3
THUSI-Lab/GameScaling-2B-S3
Tracked by OpenModelStats since October 14, 2026View on Hugging Face ↗
- Downloads 30DRolling recent-window download count as reported by the source platform.
- 493
- Total downloadsCumulative all-time download counter reported by the source platform.
- 493
- Gained 7DIncrease in cumulative total downloads between OpenModelStats' own observations seven days apart.
- —
- Gained 30DIncrease in cumulative total downloads between OpenModelStats' own observations roughly thirty days apart.
- —
- LikesCommunity likes reported by the source platform.
- 3
- Rank · tracked modelsPosition among all eligible models tracked by OpenModelStats, ordered by recent downloads. #4809 among tracked Image Text to Text models.
- #50,079
- 789(up 789 places)this week
- Parameters
- 2.7B
- Last source update
- Oct 22, 2026
- It moved from #4867 to #4809 among tracked Image Text to Text models this week.
Downloads over time
Daily gain between consecutive observationsLikes over time
Rank among tracked models
Lower is betterOverview
GameScaling-2B-S3 is a image text to text model published by THUSI-Lab. OpenModelStats has tracked the model since Oct 14, 2026, most recently observing it 8h ago.
- Published
- Oct 14, 2026
- Last updated
- Oct 22, 2026
- Library
- transformers
- License
- apache-2.0
- Task
- Image Text to Text
- Parameters
- 2,721,801,024
- Spaces
- —
- Derivative models
- —
- Velocity 7D/day
- —
transformerssafetensorsqwen3_5image-text-to-textgamescalinggame-agentembodied-agentkeyboard-mousecontinued-pretrainingconversationalenarxiv:2610.28698
Publisher
THUSI-LabView publisher statistics →Derived from Qwen/Qwen3.5-2B (reported by the source; parent not yet tracked).
Similar models
| Model | DL 30D |
|---|---|
| Qwen3-VL-8B-Instruct-quantized.w4a16SherlockID365 | 70.5K |
| Qwen3-VL-8B-Instruct-AWQ-W4A16MLliu6 | 478 |
| LFM2.5-VL-3B-OptiQ-4bitmlx-community | 755 |
| InternVL3_5-8B-AWQ-4bitcyankiwi | 185 |
| Sa2VA-Qwen3-VL-2BByteDance | 326 |
| Qwen3.5-ocr-jp-2bebinan92 | 303 |