ByteDance: UI-TARS 7B
ByteDance UI-TARS 7B is a small vision model with a 128K context window at a very low price, but without benchmark data its capabilities cannot be assessed — it appears targeted at UI understanding tasks rather than general business use.
Assessment date: March 12, 2026
Our methodology takes into account a range of factors including pricing, functionality, capabilities, benchmark performance, and real-world applicability. Rankings are reviewed and updated regularly as new models are released. Issues with our rankings? Contact us
UI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games. Built by ByteDance, it builds upon the UI-TARS framework with reinforcement learning-based reasoning, enabling robust action planning and execution across virtual interfaces. This model achieves state-of-the-art results on a range of interactive and grounding benchmarks, including OSworld, WebVoyager, AndroidWorld, and ScreenSpot. It also demonstrates perfect task completion across diverse Poki games and outperforms prior models in Minecraft agent tasks. UI-TARS-1.5 supports thought decomposition during inference and shows strong scaling across variants, with the 1.5 version notably exceeding the performance of earlier 72B and 7B checkpoints.
Capabilities
Architecture
| Modality | Text + Image → Text |
| Tokenizer | Other |
| Parameters | 7B |
Model Information
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.10 | $0.000100 |
| Output | $0.20 | $0.000200 |
Live Performance
Live endpoint metrics — refreshed every 30 minutes.
External Resources
Data sourced from OpenRouter API, Artificial Analysis and Hugging Face Open LLM Leaderboard. Scores are editorially curated by our team.
Last updated: March 13, 2026 7:52 pm