Downloads · 30 days
12
18% of all-time downloads
Janus-Pro-7B-GCPO
KonstantinosKK/Janus-Pro-7B-GCPO
Janus-Pro-7B-GCPO is a text-to-image model from KonstantinosKK. Use it when you need an image from a text prompt. It is set up for transformers. The card lists the license as other.
Github · Paper · arXiv · Checkpoints · Website
All-time downloads
65
Public
Parameters
7.4B
14.8 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors14.8 GB · 100%
At a glance
- Task
- Text-to-Image
- Library
- transformers
- License
- other
- Model type
- multi_modality
- Access
- Public
- Created
- May 27, 2026
- Updated
- May 28, 2026
- SHA
- f84c1165
Base models
Model card
From the Hugging Face model README
GCPO Janus-Pro-7B (Transformers-Compatible)
Github · Paper · arXiv · Checkpoints · Website
Janus-Pro-7B finetuned with Guidance Contrastive Policy Optimization (GCPO), a per-token credit assignment method for GRPO-style RL. Each token's advantage is weighted by the KL divergence between the policy's predictions under a positive vs. negative prompt — using the classifier-free guidance signal as a token saliency map.
<p align="center"> <img src="assets/explainer_gcpo.png" width="900" alt="GCPO Explainer"/> </p>Usage
Please refer to the Janus Github Repository.
License
The use of Janus-Pro models is subject to the DeepSeek Model License.
- Task
- Text-to-Image
- Library
- transformers
- Type
- multi_modality
- License
- other
- Languages
- en
- Created
- May 27, 2026
- Updated
- May 28, 2026