The Known Good Updated 29 Jul 2026
Alibaba ·Proprietary ·Released August 2025·Tracked since 25 Jul 2026

Qwen: Qwen3 VL 235B A22B Thinking — quality, speed & price analysis

Compare with another model Provider benchmarks Data sources
Quality
unranked
—
Known Good Index
Speed
#167 of 317
46.9
Output tokens per second
Price
#230 of 370
$1.30
Blended per 1M tokens
In $0.40 · Out $4.00
Cache hit price
unranked
—
per 1M tokens
Context
#278 of 375
131k
tokens (~200 pages)
Summary

Qwen: Qwen3 VL 235B A22B Thinking does not yet carry a Known Good Index score — the index needs at least one science, one mathematics and one coding result. It accepts text, image input, outputs text, and supports a 131k-token context window.

Pricing is $0.40 per 1M input tokens (below the $2.10 average) and $4.00 per 1M output (below the $9.62 average). At 47 tokens per second it sits in the lower middle of the tracked range.

Generated from tracked data on 29 Jul 2026 · how these comparisons are computed →

Technical specifications
ReasoningYes
Input modalitytext, image
Output modalitytext
Context window131,072 tokens
Max output32,768 tokens
WeightsProprietary
Knowledge cutoffMarch 2025
Tool callingYes
Providers2 hosts · compare →
Highlights — where Qwen: Qwen3 VL 235B A22B Thinking sits
Quality

Known Good Index · Higher is better · Composite of seven published evaluations

98
96
95
94
93
93
93
91
90
90
88
88
86
83
83
80
80
78
78
77
75
75
74
74
74
73
72
70
68
68
gpt-5.5-pre-release
Gemini 3.1 Pro
Preview
Gemini 3.5 Flash
Claude Opus 4.7
Qwen3.7 Max
DeepSeek V4 Pro
Kimi K2.6
GLM 5.2
Qwen3.6 Max
Preview
GPT-5.4
Claude Opus 4.6
GLM 5.1
o3 Mini
Gemini 3 Flash
Preview
gemini-3-pro-preview
GPT-5.2
Claude Sonnet 4.6
GPT-5
Qwen3.6 Plus
GLM 5
R1
deepseek-r1
claude-3-7-sonnet-20250219
GPT-5.1
o3
gemini-2.0-pro-exp-02-05
claude-opus-4-5-20251101
o1
claude-opus-4-20250514
GLM 4.7
Speed

Output tokens per second · Higher is better · OpenRouter endpoint telemetry

2,504
1,909
1,434
566
370
342
280
181
166
161
154
148
145
144
143
136
135
135
134
132
131
130
130
126
126
126
124
121
108
107
47
Morph V3 Large
Morph V3 Fast
Relace Apply 3
Grok 4.20
Multi-Agent
gpt-oss-safeguard-20b
Nano Banana 2 Lite
(Gemini 3.1 Flash Lite Image)
Mercury 2
Nano Banana
(Gemini 2.5 Flash Image)
Grok 4.20
Nano Banana 2
(Gemini 3.1 Flash Image Preview)
Llama 3.2 3B
Instruct
Qwen3 Next 80B A3B
Thinking
gpt-oss-120b
Gemini 3.6 Flash
Laguna XS 2.1
Muse Spark 1.1
Trinity Large
Thinking
o3 Mini High
Qwen3 30B A3B
Thinking 2507
gpt-oss-20b
Nano Banana 2
(Gemini 3.1 Flash Image)
Gemini 3.1 Flash
Lite Preview
GPT-5.6 Luna Pro
Qwen3 VL 8B
Thinking
Gemini 3.5 Flash
Llama 3.2 1B
Instruct
Gemini 3.1 Flash
Lite
Grok Build 0.1
Claude Opus 4.8
(Fast)
Nemotron 3 Nano
30B A3B
Qwen3 VL 235B A22B
Thinkingthis model #167/317
Price

Blended USD per 1M tokens · Lower is better · Published pricing via OpenRouter

$0.01
$0.02
$0.04
$0.04
$0.04
$0.06
$0.06
$0.06
$0.06
$0.06
$0.06
$0.06
$0.06
$0.06
$0.07
$0.07
$0.07
$0.07
$0.07
$0.07
$0.07
$0.08
$0.09
$0.09
$0.09
$0.10
$0.10
$0.10
$0.10
$0.11
$1.30
Ling-2.6-flash
Mistral Nemo
Granite 4.0 Micro
Llama 3 8B Lunaris
Nex-N2-Mini
Qwen2.5 7B
Instruct
gpt-oss-20b
Qwen3.7 Flash
Mistral Small 3
Llama 3.1 8B
Instruct
MythoMax 13B
Nova Micro 1.0
Granite 4.1 8B
Gemma 3 4B
Command R7B
(12-2024)
GPT-5 Nano (batch)
gpt-oss-120b
Llama 3.2 1B
Instruct
Gemma 3 12B
Laguna XS 2.1
Gemma 3n 4B
Qwen3 30B A3B
Instruct 2507
Nemotron 3 Nano
30B A3B
Gemini 2.5 Flash Lite
(batch)
Phi 4
Hy3 preview
Ministral 3 3B
2512
Reka Edge
Nova Lite 1.0
Qwen3.5-9B
Qwen3 VL 235B A22B
Thinkingthis model #230/370 · off scale
Jump to section

Quality

How Qwen: Qwen3 VL 235B A22B Thinking compares against every tracked model

Known Good IndexCapability indices
Known Good Index

Composite of GPQA Diamond, Mock AIME 2024–25, MATH Level 5, SWE-bench Verified, LiveBench, Humanity's Last Exam and Terminal-Bench, normalised 0–100. Scores are ingested from their publishers; we do not run these evaluations.

⇩
98
96
95
94
93
93
93
91
90
90
88
88
86
💡 A lightbulb marks a model its provider publishes as a reasoning model
The Known Good

Quality Breakdown

Qwen: Qwen3 VL 235B A22B Thinking on each evaluation feeding the index, against the tracked average

Evaluation scores

Bar = Qwen: Qwen3 VL 235B A22B Thinking · grey marker = average across tracked models

None of the seven index evaluations have published a score for this model yet.

The Known Good

Arena Elo

Human preference rating from blind pairwise votes

Text Arena

Elo from blind pairwise votes, ingested from Arena

1401
Elo · 95% CI ±7
#126
of 386 ranked models
8k
votes counted
1396
style-controlled (#137)

Ingested 29 Jul 2026.

The Known Good

Price & Cost

Published API pricing by token type

BlendedInput vs outputQuality per dollar
Blended Price

USD per 1M tokens at a 3:1 input:output blend. Live from OpenRouter, cross-checked against models.dev. Lower is better.

⇩
$0.01
$0.02
$0.04
$0.04
$0.04
$0.06
$0.06
$0.06
$0.06
$0.06
$0.06
$0.06
$0.06
$1.30
The Known Good
We publish blended price rather than cost-per-task. Cost-per-task requires token counts from an evaluation suite we do not run. Blended price is computed from published rates and is fully reproducible — see the pricing methodology.

Context Window

Maximum input tokens accepted

Context windowMax output tokens
Context Window

Maximum input context from provider metadata via models.dev and OpenRouter. Higher is better.

⇩
2.00M
2.00M
2.00M
2.00M
2.00M
1.31M
1.05M
1.05M
1.05M
1.05M
1.05M
1.05M
131k
The Known Good

Speed

Median output throughput across hosting providers

Output speedBy provider
Output Speed

Median output tokens per second across all tracked providers. Derived from OpenRouter endpoint telemetry. Higher is better.

⇩
2,504
1,909
1,434
566
370
342
280
181
166
161
154
148
47
The Known Good

Latency

Time to first token, median across providers

Latency (TTFT)

Seconds to first streamed chunk, median across tracked providers. Derived from OpenRouter endpoint telemetry. Lower is better.

⇩
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.3
0.3
4.0
The Known Good

Providers

Hosts serving Qwen: Qwen3 VL 235B A22B Thinking, with their own price and speed

2 providers
Full provider benchmarks →
ProviderInput $/MOutput $/M BlendedSpeedTTFTContext
Alibaba · first party $0.40 $4.00 $1.30 58 0.6s 131k
Novita $0.98 $3.95 $1.72 40 0.8s 131k

Per-host pricing and telemetry from OpenRouter. Speed is output tokens per second; TTFT is time to first token.

The Known Good

History

How price and quality have moved since we began tracking this model

Quality & price over time

Since 25 Jul 2026

Known Good Index —
Blended price $1.30 per 1M · observed 29 Jul 2026
Output speed 47 tok/s · observed 29 Jul 2026
Time to first token 4.0s
28 Jul 2026 · price change
Qwen: Qwen3 VL 235B A22B Thinking price up 54%
Blended 3:1 price moved from $0.84 to $1.30 per 1M tokens (OpenRouter).
The Known Good

Comparisons

Head-to-head pages pairing Qwen: Qwen3 VL 235B A22B Thinking with other tracked models

Head-to-head pages are generated for the top 50 models by Known Good Index, which Qwen: Qwen3 VL 235B A22B Thinking is not currently in. Its nearest tracked models by index are below.

Browse all comparisons →
Sources for this page
Pricing & providers — OpenRouter Metadata — models.dev Elo — Arena Full attribution →