Skip to content
ModelCap

Qwen3 VL 235B A22B Thinking

Current board #45Arena #142Open weights

Qwen·qwen/qwen3-vl-235b-a22b-thinking

Qwen3 VL 235B A22B Thinking is a language model published by Qwen. 2 OpenRouter endpoints are published in the latest snapshot. It accepts text, image input and returns text. The published context limit is 131K tokens. The catalogue declares support for reasoning, tool use, structured outputs.

ModelCap Index · Current board
61.7#45
Measured · 73% support · score interval 61.7–70.1
Market Gravity
15.2
Secondary market signal
Input
$0.40
per 1M tokens
Output
$4.00
per 1M tokens
Context
131K
tokens
Providers
2
endpoints
Gateway spend
—
latest · Vercel share

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
Novitabf16Cheapest$0.98$3.95131Kbf16100.0%
Alibabafp8$0.40$4.00131Kfp8—

Hugging Face Inference Providers

Exact repository identity

Provider economics and telemetry published by the Hugging Face Router. These are operational signals, not benchmark results.

ProviderInputOutputContextTTFTThroughputFeatures
novita
live
$0.98$3.95131K854 ms40.9 tok/stools
featherless-ai
live
——————

ModelCap Index · Current board

Methodology
61.7
Measured · 2 public benchmark observations across 2 boards
Current board
#45
Public rank basis
Measured (measured)
Index support
72.7% · strong
Index interval
61.7–70.1
Rank posterior
Not available
Top 5 / Top 10 probability
Not available / Not available
Posterior as of
Snapshot timestamp unavailable
Identity binding
Exact catalogue product · qwen/qwen/qwen3-vl-235b-a22b-thinking · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata Qwen/Qwen3-VL-235B-A22B-Thinking@6664affde68449468deb7527186455c7450c13c0 (not the evaluation revision)
Direct benchmark observations
2
Breadth-adjusted BES point estimate
65.9
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
65.9
Evidence status
confirmed · 80% mass
General preference
65.2
Coding
68.7
Agents & tools
—
Reasoning
—
Evidence breadth
92%
Evidence coverage
73%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
3.9
Liquidity (providers × uptime)
39.7
Open reach (HF downloads)
11.0
Surface (context / tools / modalities)
85.2
Freshness
29.8
  • · OpenRouter popularity #239
  • · 2 live providers
  • · HF 30d downloads 12194
  • · Catalogue: context 131072, reasoning, tools, modalities Text/Image
  • · First seen 2025-09-23

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

BES evidence selected · measured input only

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.

Deployment readiness

Metadata index
83.3
strong · 94% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
100.0
Deployability
93.5
Evaluation provenance
48.9

Missing: artifactReproducibility.base-lineage, 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
15.2
OpenRouter weekly popularity
#239 of 298

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

Status
Ranked
Source
Arena
Official rank
#142 of 385
Rating
1395.4
95% confidence interval
1389–1402
Votes
8K
Source snapshot
6 Aug 2026, 13:00 UTC
Category
overall
Identity match confidence
100%

Market Gravity breakdown

Usage55%
3.9

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
39.7

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

Open reach15%
11.0

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

Freshness5%
29.8

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

Specification

Context window
131K tokens
Max output
33K tokens
Inputs
Text, Image
Outputs
Text
Tokenizer
Qwen3
Cached input tokens
Not offered
First seen on OpenRouter
23 Sept 2025
Knowledge cutoff
2025-03-31

Capabilities

  • Supported: Reasoning
  • Supported: Tool use
  • Supported: Structured output
  • 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
6664affde684
Parameters
235.7B
Architecture
Qwen3VLMoeForConditionalGeneration
Model type
qwen3_vl_moe
Hugging Face downloads (30d)
12K
Downloads (all time)
3.5M
Likes
400
Repository
Qwen/Qwen3-VL-235B-A22B-Thinking
Repository updated
26 Nov 2025, 13:18 UTC