Skip to content
ModelCap

Qwen3 235B A22B

Current board #49Arena #165Open weights

Qwen·qwen/qwen3-235b-a22b

Qwen3 235B A22B is a language model published by Qwen. 1 OpenRouter endpoint is published in the latest snapshot. It accepts text input and returns text. The published context limit is 131K tokens. The catalogue declares support for reasoning, tool use.

ModelCap Index · Current board
58.2#49
Measured · 83% support · score interval 58.2–64.8
Market Gravity
17.7
Secondary market signal
Input
$0.455
per 1M tokens
Output
$1.82
per 1M tokens
Context
131K
tokens
Providers
1
endpoint
Gateway spend
—
latest · Vercel share

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
Alibabafp8$0.455$1.82131Kfp8100.0%

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
nscale
live
$0.20$0.6032K729 ms15.8 tok/stools · structured
novita
live
$0.20$0.8041K747 ms30.0 tok/s—
featherless-ai
live
——————

Reasoning efforts

Best supported result can inform ModelCap

Published Max, X-High, High and other configurations for this model. These are sourced scores, not head-to-head essays.

Related community configurations

Human preference rating
ConfigurationRatingSource rankVotesSource
No thinking
qwen3-235b-a22b-no-thinking
1403
CI 1398–1407
#13238KArena
Match 100%

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
ScaleAI/SWE-bench_Pro
SWE_Bench_Pro
—21.41
Unpinned
not published
Reported
Quarantined
TIGER-Lab/MMLU-Pro
mmlu_pro
—68.18
Unpinned
not published
Reported
Quarantined

ModelCap Index · Current board

Methodology
58.2
Measured · 2 public benchmark observations across 2 boards
Current board
#49
Public rank basis
Measured (measured)
Index support
83.4% · strong
Index interval
58.2–64.8
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 · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata Qwen/Qwen3-235B-A22B@8efa61729e24bd65b1d152b5ab5409052aa80e65 (not the evaluation revision)
Direct benchmark observations
2
Breadth-adjusted BES point estimate
61.5
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
61.5
Evidence status
confirmed · 83% mass
General preference
60.2
Coding
67.0
Agents & tools
—
Reasoning
—
Evidence breadth
92%
Evidence coverage
83%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
7.4
Liquidity (providers × uptime)
16.2
Open reach (HF downloads)
58.1
Surface (context / tools / modalities)
70.2
Freshness
17.0
  • · OpenRouter popularity #195
  • · 1 live provider
  • · HF 30d downloads 609764
  • · Catalogue: context 131072, reasoning, tools
  • · First seen 2025-04-28

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
85.0
strong · 94% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
100.0
Deployability
96.3
Evaluation provenance
52.8

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
17.7
OpenRouter weekly popularity
#195 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
#165 of 385
Rating
1374.8
95% confidence interval
1370–1379
Votes
26K
Source snapshot
6 Aug 2026, 13:00 UTC
Category
overall
Identity match confidence
100%

Market Gravity breakdown

Usage55%
7.4

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
16.2

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

Open reach15%
58.1

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

Freshness5%
17.0

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

Specification

Context window
131K tokens
Max output
8K tokens
Inputs
Text
Outputs
Text
Tokenizer
Qwen3
Cached input tokens
Not offered
First seen on OpenRouter
28 Apr 2025
Knowledge cutoff
2025-03-31

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
8efa61729e24
Parameters
235.1B
Architecture
Qwen3MoeForCausalLM
Model type
qwen3_moe
Hugging Face downloads (30d)
610K
Downloads (all time)
6.4M
Likes
1K
Repository
Qwen/Qwen3-235B-A22B
Repository updated
26 Jul 2025, 03:45 UTC