Skip to content
ModelCap

UI-TARS 7B

Official publisher releaseOpen weights

ByteDance·bytedance/ui-tars-1.5-7b

UI-TARS 7B listed OpenRouter input price is $0.10 per 1M tokens and listed output price is $0.20 per 1M tokens, from OpenRouter listed prices as of 22 Sept 2026, 02:49 UTC.

UI-TARS 7B has a published context limit of 128K tokens in the ModelCap catalogue as of 22 Sept 2026, 02:49 UTC.

UI-TARS 7B has 1 published OpenRouter serving endpoint, from OpenRouter as of 22 Sept 2026, 02:49 UTC.

UI-TARS 7B weight access is open weights, from ModelCap classification of published repository metadata as of 22 Sept 2026, 02:49 UTC.

Snapshot facts · download the public dataset · methodology

ModelCap Index · Current board
Not scored · 0% support · score interval not available
Market Gravity
19.1
Secondary market signal
Input
$0.10
per 1M tokens
Output
$0.20
per 1M tokens
Context
128K
tokens
Providers
1
endpoint
Gateway spend
latest · Vercel share

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
Parasailbf16$0.10$0.20128Kbf16100.0%

Reported evaluation leads (27)

Not independent measurement

Structured rows reported in this exact model repository at the pinned revision below. They are the publisher talking about itself: never measured evidence, never a score input on their own. They remain quarantined until dataset revision, harness, configuration and identity pass ModelCap’s independent admission review.

Dataset / taskMetricValueRevisionProvenance
likaixin/ScreenSpot-Pro
overall
61.6
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
android_studio_macos
57.5
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
autocad_windows
41.2
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
blender_windows
59.2
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
davinci_macos
54.5
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
eviews_windows
96
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
excel_macos
68.8
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
fruitloops_windows
43.9
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
illustrator_windows
22.6
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
inventor_windows
58.6
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
linux_common_linux
66
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
macos_common_macos
56.9
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
matlab_macos
81.7
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
origin_windows
45.2
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
photoshop_windows
56.9
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
powerpoint_windows
79.3
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
premiere_windows
42.3
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
pycharm_macos
65.4
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
quartus_windows
44.4
Unpinned
not published
Reported
Quarantined
likaixin/ScreenSpot-Pro
solidworks_windows
53.2
Unpinned
not published
Reported
Quarantined

Showing 20 of 27 rows.

ModelCap Index · Current board

Methodology
Not scored · peers unresolved
Public rank basis
Not scored (abstained)
Index support
0.0% · not applicable
Index interval
Not available
Rank posterior
Not available
Top 5 / Top 10 probability
Not available / Not available
Posterior as of
Snapshot timestamp unavailable
Evidence as of
No admitted evaluation publication timestamp
Identity binding
Exact catalogue product · bytedance/bytedance/ui-tars-1.5-7b · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata ByteDance-Seed/UI-TARS-1.5-7B@683d002dd99d8f95104d31e70391a39348857f4e (not the evaluation revision)
Index observations used
0
Architecture fallback
abstained · 0% coverage · 0 anchors
Bayesian Evidence Score · measured evidence input only

BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.

Observed capability
Evidence status
No admitted measured evidence
General preference
Coding
Agents & tools
Reasoning
Evidence breadth
0%
Evidence coverage
0%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
3.6
Liquidity (providers × uptime)
20.3
Open reach (HF downloads)
73.6
Surface (context / tools / modalities)
40.0
Freshness
19.7
  • · OpenRouter popularity #261
  • · 1 live provider
  • · HF 30d downloads 600259
  • · Catalogue: context 128000, modalities Image/Text
  • · First seen 2025-07-22

These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.

The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.

Experimental capability estimate

Estimate policy
15.4
0.0–32.1 uncertainty interval
Excluded from rank
Confidence
71%
Training support
209 models · 120 lineages
Out-of-domain check
in domain
Feature coverage
100%
Lineage-held-out error
10.0 MAE

Experimental estimate. This layer predicts from admitted evidence and safe metadata only. It never enters ModelCap Score or rank.

Deployment readiness

Metadata index
67.7
partial · 76% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
100.0
Deployability
74.0
Evaluation provenance
0.0

Missing: artifactReproducibility.base-lineage, evaluationProvenance.admitted-evaluation-observations, evaluationProvenance.capability-family-breadth, evaluationProvenance.evaluation-source-diversity, evaluationProvenance.identity-and-source-confidence, evaluationProvenance.evaluation-dataset-revisions.

A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.

Market activity

Venue coverage
Market Gravity
19.1
OpenRouter weekly popularity
#261 of 322

Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.

Arena preference

No matched Arena rating

This model is visible because it is available in the live catalogue, but it has no trustworthy community result yet. ModelCap does not infer a quality score from price, features, or market activity.

Market Gravity breakdown

Usage55%
3.6

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
20.3

Independent providers versus the model's open or closed cohort, weighted by uptime.

Open reach15%
73.6

Hugging Face 30-day downloads for open models; neutral for closed models.

Freshness5%
19.7

Time since first public availability, on a six-month half-life.

Specification

Context window
128K tokens
Max output
2K tokens
Inputs
Image, Text
Outputs
Text
Tokenizer
Other
Cached input tokens
$0.10 / 1M
First seen on OpenRouter
22 Jul 2025
Knowledge cutoff
2025-01-31

Capabilities

  • Not supported: Reasoning
  • Not supported: Tool use
  • Supported: Structured output
  • Not supported: Response format
  • Not supported: Moderated

Weights & access

Open weights. Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI.

Access
Open weights
Licence
apache-2.0
Pinned revision
683d002dd99d
Parameters
8.3B
Architecture
Qwen2_5_VLForConditionalGeneration
Model type
qwen2_5_vl
Hugging Face downloads (30d)
600K
Downloads (all time)
4.3M
Likes
605
Repository updated
18 Apr 2025, 01:35 UTC

UI-TARS 7B: common questions

How does UI-TARS 7B rank among AI models?

UI-TARS 7B has no public ModelCap Index position as of 22 September 2026: the snapshot lacks enough matched benchmark evidence, or the identity match is unresolved. Its catalogue facts (price, context, providers) are shown without a rank.

How much does UI-TARS 7B cost per 1M tokens?

UI-TARS 7B is listed at $0.10 per 1M input tokens and $0.20 per 1M output tokens as of 22 September 2026. Prices come from the OpenRouter catalogue and re-observe on every refresh. For 1,000 requests using 2,000 input and 500 output tokens each, the listed rates imply $0.30 total (2M input + 0.5M output). This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the chosen endpoint's rate and limits before budgeting.

What is the context window of UI-TARS 7B?

UI-TARS 7B has a published context window of 128K tokens (128,000) and a maximum output of 2K tokens. Individual providers can serve less than the published maximum; the providers table lists each endpoint's own limit.

Is UI-TARS 7B open-weight?

Open weights (apache-2.0). Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI. The repository is ByteDance-Seed/UI-TARS-1.5-7B on Hugging Face.

Which API providers serve UI-TARS 7B?

One provider serves UI-TARS 7B through OpenRouter as of 22 September 2026: Parasail. Each provider's price, context limit, quantization and measured uptime are in the providers table above.

When was UI-TARS 7B released?

UI-TARS 7B first appeared in the catalogue on 22 Jul 2025, with a published knowledge cutoff of 2025-01-31. It is the current version in its family. Rank and price movements since then are recorded on the site's changes feed.