ByteDance: UI-TARS 7B
bytedance/ui-tars-1-5-7bUI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games.
Modalities
In / out per 1M$0.10 / $0.20
Context125K
ReleasedJul 22, 2025
Providers
1 provider serves this model upstream.
| Provider | Input /M | Output /M | Cache read /M | Latency P50 | Throughput P50 | Uptime |
|---|---|---|---|---|---|---|
| $0.10 | $0.20 | $0.10 | 1.11s | 44 tps | 100% |
Market-wide measurements, refreshed every 15 minutes. They describe who serves UI-TARS 7B in general, not how ProjectCOZY routes your requests.
Pricing
List rates per 1M tokens — what UI-TARS 7B costs pay-as-you-go elsewhere, not what you pay here.
Input
$0.10
per 1M tokens
Output
$0.20
per 1M tokens
Cache read
$0.10
per 1M cached tokens
Included in every pass
UI-TARS 7B is included in every ProjectCOZY pass at no extra cost — one flat rate, no per-token metering. The rates above are there for comparison.
Performance
Time to first token and output speed across every provider, over the last 3 days.
0.0s10.0s20.0s30.0s40.0s
Oct 11, 12 AMOct 11, 9 AMOct 11, 5 PMOct 12, 1 AMOct 12, 10 AMOct 12, 6 PMOct 13, 2 AMOct 13, 10 AMOct 13, 6 PMOct 14, 2 AMOct 14, 10 AM
Median 90th percentile 99th percentile
0 tok/s50 tok/s100 tok/s150 tok/s200 tok/s
Oct 11, 12 AMOct 11, 9 AMOct 11, 5 PMOct 12, 1 AMOct 12, 10 AMOct 12, 6 PMOct 13, 2 AMOct 13, 10 AMOct 13, 6 PMOct 14, 2 AMOct 14, 10 AM
Median 90th percentile 99th percentile
Uptime
100% of requests were served successfully over the last 3 days, across all providers.
Last 3 daysAvailability 99.95%
Oct 11, 2:00 PMNow
Availability over the last 24 hoursAvailability 99.95%
100% 97% 93% 90%
Oct 13, 2:50 PMNow
Apps
Tools that most often run UI-TARS 7B, by tokens this week.
Activity
1.1B tokens processed market-wide in the last month.
025.0M50.0M75.0M100.0M
Sep 14Sep 19Sep 24Sep 29Oct 4Oct 9Oct 14
Tokens
FAQ
Which model id do I use for UI-TARS 7B?
bytedance/ui-tars-1.5-7b. Send it as the model field from any OpenAI-compatible client, against http://192.168.1.55/v1.
How much context does UI-TARS 7B hold?
128,000 tokens per request, of which up to 2,048 can be the response.
What can it accept and return?
Input: image, text. Output: text.
What does UI-TARS 7B cost on a pass?
Nothing beyond the pass itself. Pay-as-you-go elsewhere it lists at $0.10 per 1M input tokens and $0.20 per 1M output tokens.
How current is its knowledge?
Training data runs to Jan 2025. Anything later has to come from your prompt or a tool call.
Who built UI-TARS 7B?
ByteDance, in the UI family.
Explore
Other models your pass can call.
