shuai bai
commited on
Update README.md
Browse files
README.md
CHANGED
|
@@ -5,28 +5,48 @@ license: apache-2.0
|
|
| 5 |
<img alt="Chat" src="https://img.shields.io/badge/%F0%9F%92%9C%EF%B8%8F%20Qwen%20Chat%20-536af5" style="display: inline-block; vertical-align: middle;"/>
|
| 6 |
</a>
|
| 7 |
|
| 8 |
-
|
| 9 |
|
| 10 |
-
The goal of **Qwen3-VL** is not just to “see” images or videos — but to truly understand the world, interpret events, and take action. To achieve this, we’ve systematically upgraded key capabilities, pushing visual models from simple “perception” toward deeper “cognition,” and from basic “recognition” to advanced “reasoning and execution.”
|
| 11 |
-

|
| 12 |
|
| 13 |
-
|
| 14 |
|
| 15 |
-
|
| 16 |
|
| 17 |
-
|
| 18 |
|
| 19 |
-
- **Greatly Improved Visual Coding**: It can now generate code from images or videos — for example, turning a design mockup into Draw.io, HTML, CSS, or JavaScript code — making “what you see is what you get” visual programming a reality.
|
| 20 |
|
| 21 |
-
|
| 22 |
|
| 23 |
-
|
| 24 |
|
| 25 |
-
|
| 26 |
|
| 27 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
|
| 29 |
-
- **Better OCR Across More Languages & Complex Scenes**: OCR now supports 32 languages (up from 10), covering more countries and regions. It performs more reliably under challenging real-world conditions like poor lighting, blur, or tilted text. Recognition accuracy for rare characters, ancient scripts, and technical terms has also improved significantly. Its ability to understand long documents and reconstruct fine structures is further enhanced.
|
| 30 |
|
| 31 |
This is the weight repository for Qwen3-VL-235B-A22B-Thinking.
|
| 32 |
|
|
|
|
| 5 |
<img alt="Chat" src="https://img.shields.io/badge/%F0%9F%92%9C%EF%B8%8F%20Qwen%20Chat%20-536af5" style="display: inline-block; vertical-align: middle;"/>
|
| 6 |
</a>
|
| 7 |
|
| 8 |
+
# Qwen3-VL-235B-A22B-Thinking
|
| 9 |
|
|
|
|
|
|
|
| 10 |
|
| 11 |
+
Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.
|
| 12 |
|
| 13 |
+
This generation delivers comprehensive upgrades across the board: superior text understanding & generation, deeper visual perception & reasoning, extended context length, enhanced spatial and video dynamics comprehension, and stronger agent interaction capabilities.
|
| 14 |
|
| 15 |
+
Available in Dense and MoE architectures that scale from edge to cloud, with Instruct and reasoning‑enhanced Thinking editions for flexible, on‑demand deployment.
|
| 16 |
|
|
|
|
| 17 |
|
| 18 |
+
#### Key Enhancements:
|
| 19 |
|
| 20 |
+
* **Visual Agent**: Operates PC/mobile GUIs—recognizes elements, understands functions, invokes tools, completes tasks.
|
| 21 |
|
| 22 |
+
* **Visual Coding Boost**: Generates Draw.io/HTML/CSS/JS from images/videos.
|
| 23 |
|
| 24 |
+
* **Advanced Spatial Perception**: Judges object positions, viewpoints, and occlusions; provides stronger 2D grounding and enables 3D grounding for spatial reasoning and embodied AI.
|
| 25 |
+
|
| 26 |
+
* **Long Context & Video Understanding**: Native 256K context, expandable to 1M; handles books and hours-long video with full recall and second-level indexing.
|
| 27 |
+
|
| 28 |
+
* **Enhanced Multimodal Reasoning**: Excels in STEM/Math—causal analysis and logical, evidence-based answers.
|
| 29 |
+
|
| 30 |
+
* **Upgraded Visual Recognition**: Broader, higher-quality pretraining is able to “recognize everything”—celebrities, anime, products, landmarks, flora/fauna, etc.
|
| 31 |
+
|
| 32 |
+
* **Expanded OCR**: Supports 32 languages (up from 19); robust in low light, blur, and tilt; better with rare/ancient characters and jargon; improved long-document structure parsing.
|
| 33 |
+
|
| 34 |
+
* **Text Understanding on par with pure LLMs**: Seamless text–vision fusion for lossless, unified comprehension.
|
| 35 |
+
|
| 36 |
+
|
| 37 |
+
#### Model Architecture Updates:
|
| 38 |
+
|
| 39 |
+
<p align="center">
|
| 40 |
+
<img src="https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3-VL/qwen3vl_arc.jpg" width="80%"/>
|
| 41 |
+
<p>
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
1. **Interleaved-MRoPE**: Full‑frequency allocation over time, width, and height via robust positional embeddings, enhancing long‑horizon video reasoning.
|
| 45 |
+
|
| 46 |
+
2. **DeepStack**: Fuses multi‑level ViT features to capture fine‑grained details and sharpen image–text alignment.
|
| 47 |
+
|
| 48 |
+
3. **Text–Timestamp Alignment:** Moves beyond T‑RoPE to precise, timestamp‑grounded event localization for stronger video temporal modeling.
|
| 49 |
|
|
|
|
| 50 |
|
| 51 |
This is the weight repository for Qwen3-VL-235B-A22B-Thinking.
|
| 52 |
|