Downloads · 30 days
17
61% of all-time downloads
VideoSSR-SelfPlay-JigsawOnly
gaon0427/VideoSSR-SelfPlay-JigsawOnly
VideoSSR-SelfPlay-JigsawOnly is a reinforcement learning model from gaon0427. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
step-300/ contains the merged actor checkpoint at global step 300 from the September 7, 2026 Jigsaw-only self-play run (qwen3vl8bjigsawprogressexploreobjectrefresume100). The run uses 48-frame videos and the fixed-v5…
All-time downloads
28
Public
Parameters
8.8B
35.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors35.1 GB · 100%
At a glance
- Task
- Reinforcement Learning
- License
- apache-2.0
- Model type
- qwen3_vl
- Access
- Public
- Created
- Aug 29, 2026
- Updated
- Oct 14, 2026
- SHA
- b77417cf
Base models
Model card
From the Hugging Face model README
VideoSSR Jigsaw Self-play
step-300/ contains the merged actor checkpoint at global step 300 from the September 7, 2026 Jigsaw-only self-play run (qwen3vl8b_jigsaw_progress_explore_objectref_resume100). The run uses 48-frame videos and the fixed-v5 circle Jigsaw-only split. The directory includes model weights and Transformers configuration/tokenizer files.
Base model: Qwen/Qwen3-VL-8B-Instruct.
- Task
- Reinforcement Learning
- Type
- qwen3_vl
- License
- apache-2.0
- Created
- Aug 29, 2026
- Updated
- Oct 14, 2026