Downloads · 30 days
18
16% of all-time downloads
Qwen3.8-27B-EXP-EVE-v1
win10/Qwen3.8-27B-EXP-EVE-v1
Qwen3.8-27B-EXP-EVE-v1 is a image-text-to-text model from win10. 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. The card lists the license as apache-2.0.
This is a successfully merged cross-architecture model, built around the popular Qwen3.8-27B as its primary backbone.
All-time downloads
114
Public
Parameters
27.8B
55.6 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors55.6 GB · 100%
At a glance
- Task
- Image-Text-to-Text
- Library
- transformers
- License
- apache-2.0
- Model type
- qwen3_5
- Access
- Public
- Created
- Aug 24, 2026
- Updated
- Aug 24, 2026
- SHA
- cbfff3b8
Try a prompt
Base models
Model card
From the Hugging Face model README
This is a successfully merged cross-architecture model, built around the popular Qwen3.8-27B as its primary backbone.
Using a Tensor Gene Evolution merging approach, I incorporated capabilities and behavioral characteristics from deepseek-ai/DeepSeek-V4-Flash-0731 and dots-studio/dots3-note-prev into the Qwen backbone.
Compared with the original base model, this merged model appears to preserve the donors' reasoning characteristics more effectively, while demonstrating broader reasoning coverage, greater depth of thought, and more diverse problem-solving behavior.
In my observations, its reasoning ability can be surprisingly strong and, in some cases, may even appear more capable than DeepSeek models. Further systematic evaluation and benchmarking are still needed to quantify these differences.
💖 Open Sponsorship
Support this research and future model development:
- Task
- Image-Text-to-Text
- Library
- transformers
- Type
- qwen3_5
- License
- apache-2.0
- Languages
- en, zh
- Created
- Aug 24, 2026
- Updated
- Aug 24, 2026