Skip to content
ModelCap

Qwen3 235B A22B Thinking 2507

Current board #44Arena #138Open weights

Qwen·qwen/qwen3-235b-a22b-thinking-2507

Qwen3 235B A22B Thinking 2507 is a language model published by Qwen. 4 OpenRouter endpoints are published in the latest snapshot. It accepts text input and returns text. The published context limit is 262K tokens. The catalogue declares support for reasoning, tool use.

ModelCap Index · Current board
62.1#44
Measured · 76% support · score interval 62.1–69.9
Market Gravity
25.0
Secondary market signal
Input
$0.23
per 1M tokens
Output
$2.30
per 1M tokens
Context
262K
tokens
Providers
4
endpoints
Gateway spend
—
latest · Vercel share

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
DeepInfrafp8Cheapest$0.23$2.30262Kfp8100.0%
Alibabafp8$0.23$2.30131Kfp8100.0%
Novitafp8$0.30$3.00131Kfp893.62%
Venicefp8$0.45$3.50128Kfp8—

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
deepinfra
live
$0.23$2.30262K343 ms45.2 tok/stools
novita
live
$0.30$3.00131K1170 ms64.2 tok/stools
featherless-ai
live
——————

Reported evaluation leads (2)

Excluded from score

Structured rows reported in this exact model repository at the pinned revision below. They are useful research leads, but remain quarantined until dataset revision, harness, configuration and identity pass ModelCap’s independent admission review.

Dataset / taskMetricValueRevisionProvenance
LEXam-Benchmark/LEXam
open_question
—47.25
Unpinned
not published
Reported
Quarantined
LEXam-Benchmark/LEXam
mcq_4_choices
—48.19
Unpinned
not published
Reported
Quarantined

ModelCap Index · Current board

Methodology
62.1
Measured · 2 public benchmark observations across 2 boards
Current board
#44
Public rank basis
Measured (measured)
Index support
76.3% · strong
Index interval
62.1–69.9
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/qwen3-235b-a22b-thinking-2507 · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata Qwen/Qwen3-235B-A22B-Thinking-2507@6cbffae6d8e28b986a6b17bd36f42f9fa0f1f0a5 (not the evaluation revision)
Direct benchmark observations
2
Breadth-adjusted BES point estimate
66.0
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
66.0
Evidence status
confirmed · 81% mass
General preference
66.0
Coding
65.7
Agents & tools
—
Reasoning
—
Evidence breadth
92%
Evidence coverage
76%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
11.5
Liquidity (providers × uptime)
59.1
Open reach (HF downloads)
18.4
Surface (context / tools / modalities)
75.3
Freshness
23.7
  • · OpenRouter popularity #155
  • · 4 live providers
  • · HF 30d downloads 31424
  • · Catalogue: context 262144, reasoning, tools
  • · First seen 2025-07-25

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
82.6
strong · 94% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
100.0
Deployability
90.7
Evaluation provenance
49.2

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
25.0
OpenRouter weekly popularity
#155 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
#138 of 385
Rating
1398.9
95% confidence interval
1392–1405
Votes
9K
Source snapshot
6 Aug 2026, 13:00 UTC
Category
overall
Identity match confidence
100%

Market Gravity breakdown

Usage55%
11.5

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
59.1

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

Open reach15%
18.4

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

Freshness5%
23.7

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

Specification

Context window
262K tokens
Max output
—
Inputs
Text
Outputs
Text
Tokenizer
Qwen3
Cached input tokens
Not offered
First seen on OpenRouter
25 Jul 2025
Knowledge cutoff
2025-06-30

Capabilities

  • Supported: Reasoning
  • Supported: Tool use
  • Not 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
6cbffae6d8e2
Parameters
235.1B
Architecture
Qwen3MoeForCausalLM
Model type
qwen3_moe
Hugging Face downloads (30d)
31K
Downloads (all time)
624K
Likes
408
Repository
Qwen/Qwen3-235B-A22B-Thinking-2507
Repository updated
17 Aug 2025, 08:20 UTC