Hugging Face's logo

bytedance-research
/
UI-TARS-2B-SFT

Image-Text-to-Text
Transformers
Safetensors
English
qwen2_vl
multimodal
gui
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
2
MingComplex commited on
Commit
ff639e9
·
2 Parent(s): b0fe7fc d47e14f

update readme

Browse files
Files changed (1) hide show
  1. README.md +1 -4
README.md CHANGED
@@ -17,7 +17,7 @@ library_name: transformers
17
  [**UI-TARS-7B-DPO**](https://huggingface.co/bytedance-research/UI-TARS-7B-DPO)(Recommended)  | 
18
  [UI-TARS-7B-gguf](https://huggingface.co/bytedance-research/UI-TARS-7B-gguf)  | 
19
  [UI-TARS-72B-SFT](https://huggingface.co/bytedance-research/UI-TARS-72B-SFT)  | 
20
- [**UI-TARS-72B-DPO**](https://huggingface.co/bytedance-research/UI-TARS-72B-DPO)(Recommended)
21
  ## Introduction
22
 
23
  UI-TARS is a next-generation native GUI agent model designed to interact seamlessly with graphical user interfaces (GUIs) using human-like perception, reasoning, and action capabilities. Unlike traditional modular frameworks, UI-TARS integrates all key components—perception, reasoning, grounding, and memory—within a single vision-language model (VLM), enabling end-to-end task automation without predefined workflows or manual rules.
@@ -32,9 +32,6 @@ UI-TARS is a next-generation native GUI agent model designed to interact seamles
32
  <!-- ![Local Image](figures/UI-TARS-vs-Previous-SOTA.png) -->
33
 
34
  This repository contains the model for the paper [UI-TARS: Pioneering Automated GUI Interaction with Native Agents](https://huggingface.co/papers/2501.12326).
35
-
36
- Code: https://github.com/bytedance/UI-TARS
37
-
38
  ## Performance
39
  **Perception Capabilty Evaluation**
40
  | Model | VisualWebBench | WebSRC | SQAshort |
 
17
  [**UI-TARS-7B-DPO**](https://huggingface.co/bytedance-research/UI-TARS-7B-DPO)(Recommended) &nbsp;|&nbsp;
18
  [UI-TARS-7B-gguf](https://huggingface.co/bytedance-research/UI-TARS-7B-gguf) &nbsp;|&nbsp;
19
  [UI-TARS-72B-SFT](https://huggingface.co/bytedance-research/UI-TARS-72B-SFT) &nbsp;|&nbsp;
20
+ [UI-TARS-72B-DPO](https://huggingface.co/bytedance-research/UI-TARS-72B-DPO)
21
  ## Introduction
22
 
23
  UI-TARS is a next-generation native GUI agent model designed to interact seamlessly with graphical user interfaces (GUIs) using human-like perception, reasoning, and action capabilities. Unlike traditional modular frameworks, UI-TARS integrates all key components—perception, reasoning, grounding, and memory—within a single vision-language model (VLM), enabling end-to-end task automation without predefined workflows or manual rules.
 
32
  <!-- ![Local Image](figures/UI-TARS-vs-Previous-SOTA.png) -->
33
 
34
  This repository contains the model for the paper [UI-TARS: Pioneering Automated GUI Interaction with Native Agents](https://huggingface.co/papers/2501.12326).
 
 
 
35
  ## Performance
36
  **Perception Capabilty Evaluation**
37
  | Model | VisualWebBench | WebSRC | SQAshort |