Downloads · 30 days
6
9% of all-time downloads
qwen35-4b-dpo-grove-p
Sarim-Hash/qwen35-4b-dpo-grove-p
qwen35-4b-dpo-grove-p is a image-text-to-text model from Sarim-Hash. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for transformers.
Full-parameter GRPO policy with the process reward, used as q4b-g2-grove-p in the ICLR persuasion debate experiments.
All-time downloads
66
Public
Parameters
5.2B
10.4 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors10.4 GB · 100%
At a glance
- Task
- Image-Text-to-Text
- Library
- transformers
- Model type
- qwen3_5
- Access
- Public
- Created
- Aug 30, 2026
- Updated
- Aug 30, 2026
- SHA
- 933a8425
Try a prompt
Base models
Model card
From the Hugging Face model README
Qwen3.5-4B DPO + GROVE-P
Full-parameter GRPO policy with the process reward, used as q4b-g2-grove-p in the ICLR persuasion debate experiments.
- Initialization: Winner-v2 DPO
- Artifact: evaluated
finalcheckpoint after 3,000 GRPO steps - Weight file:
model.safetensors - SHA-256:
7aa2bac9c8469a0a880606a514a054be2e429903b235ebc0165fd8957d748743 - This is a full fine-tune, not a LoRA adapter.
- Task
- Image-Text-to-Text
- Library
- transformers
- Type
- qwen3_5
- Created
- Aug 30, 2026
- Updated
- Aug 30, 2026