UI TARS 7B
1351°
Run this model
Offers / Ads
ByteDance | 128,000 token context | $0.10/M input, $0.20/M output. 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…ByteDance | 128,000 token context | $0.10/M input, $0.20/M output. 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.
Guides for this model
Ollama vs LM Studio (2026): which local AI app to use, and which top models each one runs
How much VRAM do I actually need for local AI?
Open weights vs open source vs API-only: what you are actually allowed to do
What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)
Captain's notes
Notes from 2026-10-13
UI-TARS-1.5 7B is ByteDance's open-weights multimodal agent built to operate graphical interfaces: desktop apps, browsers, mobile screens and games. Reinforcement-learning-based reasoning lets it plan and execute multi-step actions rather than just describe what it sees. At 7B parameters it punches well above its weight class, outperforming earlier 72B checkpoints on several GUI benchmarks.
Known for
- Top scores on OSworld, AndroidWorld and ScreenSpot grounding benchmarks
- Operates desktop, browser, mobile and game interfaces as an autonomous agent
- Thinks through steps before acting, thanks to RL-trained reasoning
- Apache 2.0 licence — run it locally, modify it, deploy it commercially
- 7B size outperforms the maker's own earlier 72B model on key tasks
Watch out for
- Primarily English; multilingual GUI support is not documented by ByteDance.
- Running it locally as an agent requires more setup than a chat model — you need a GUI harness.
- Context is 128k tokens, but agent tasks that loop over many screenshots can burn through it quickly.
Prompts to try
Browser task automation
You are controlling a web browser. Go to Wikipedia, search for 'reinforcement learning', find the section on applications, and return the first sentence of that section. Describe every click and keystroke you take, then give me the final answer.
Screen element grounding
I am attaching a screenshot of a desktop application. Identify the exact pixel coordinates of the primary action button, explain how you located it, and describe what action you would take next to complete a file-save operation.
Multi-step mobile task
You are operating an Android phone. Starting from the home screen, open the Settings app, navigate to Wi-Fi, and toggle it off. List each tap or swipe as a numbered step, and flag any point where the UI might require a confirmation dialog.
Compare with
- GPT-5.6 LunaAPI-only model1,050,000 token context · $0.20/M input, $1.20/M output
- Muse Spark 1.3 ContributorAPI-only model1,048,576 token context · $0.10/M input, $0.20/M output
- Claude Haiku 5.5API-only model1,000,000 token context · $0.10/M input, $0.50/M output
- Gemini 2.5 Flash LiteAPI-only model1,048,576 token context · $0.10/M input, $0.40/M output
More text models from bytedance
Questions people ask
- Can I run UI-TARS-1.5 7B locally?
- Yes — the weights are open under Apache 2.0 and hosted on Hugging Face, so you can download and run them on your own hardware. ByteDance also publishes a desktop application wrapper on GitHub to make local agent use easier.
- What does it cost to use via API, and what is the context window?
- Pricing is $0.10 per million input tokens and $0.20 per million output tokens, with a 128,000-token context window.
- How does UI-TARS-1.5 7B differ from the original UI-TARS models?
- The 1.5 version adds reinforcement-learning-based reasoning, which lets the model decompose tasks and plan actions before executing them; ByteDance reports it outperforms both the earlier 7B and 72B checkpoints on benchmark tasks.
Captain's verdictA rare 7B model that earns its place in serious agent pipelines — open weights, low cost, and genuinely competitive with much larger closed rivals on GUI tasks.
Model card on Hugging Face
Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.
Ranking badge
Updated Oct 22, 2026
Made this model? Show this badge on your page.
Paste it into your Hugging Face card or Civitai description.
How this score adds up
This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.
Glossary: the terms used here
- Heat score
- One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
- Floor
- The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
- Download-equivalents
- Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
- Log scale
- Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
- Likes
- Thumbs up and likes from every source, worth half a download each on the log scale.
- Adoption rate
- How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
- Measured and estimated
- Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
- Seeder
- Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
- AI Horde
- A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
- Generation sites
- Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
- OpenRouter tokens
- A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
- Full model (checkpoint)
- The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
- LoRA and embedding
- Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
- Family and weights
- The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
- Age
- Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
- Family boost
- A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
- Stand-in score
- A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
- New! (grace period)
- A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
- Trend
- How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
- Typical model
- The middle of the catalog: half the models with a figure grew faster over the period, half slower.
UI TARS 7B is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.
Browse by Model Family
Text / LLM Models
434
Anima Models
528
Chroma Models
19
Fibo Models
2
Flux Models
1,496
Flux 2 / Klein Models
1,239
Flux 3 Models
3
Moss Models
3
Iris Models
1
MiniMax H3 Models
364
Hunyuan Models
204
Ideogram Models
20
Krea2 Models
717
Qwen Models
20
Qwen2 Models
47
Ltx2 Models
153