ByteDance: UI-TARS 7B

by Bytedancetext+image->text1 endpoint

Overview

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...

Capabilities

Text generation
Image understanding
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
ImageText
Output
Text

Technical Specifications

Context Window
128K tokens
Max Output
2.0K tokens
Knowledge Cutoff
2025-01-31
Tokenizer
Other
Uptime (24h)
10000.0%

Supported Parameters

frequency_penaltylogit_biaslogprobsmax_tokenspresence_penaltyrepetition_penaltyseedstopstructured_outputstemperaturetop_ktop_logprobstop_p

Pricing

Per 1M tokens
Input$0.10
Output$0.20
Cache read$0.10
Added
July 22, 2025
Last synced 9/16/2026