Qwen: Qwen3 235B A22B Instruct 2507

qwen · Released Jul 21, 2025
DFO score #151 / 743
18.4 leaderboard score
Speed #262 / 353
59.3 tok/s
Input Price #312 / 743
$0.140 per 1M tokens
Output Price #386 / 743
$0.800 per 1M tokens
Context #202 / 743
262,144 tokens

Analysis Summary

Qwen3 235B A22B Instruct 2507 is a text model with a 262K context window, tool use, and function calling. Its supplied math and coding benchmark results are strong, and its agentic result is meaningful, though its overall intelligence index remains below the highest-capability models. Listed pricing is $0.0875 per million input tokens and $0.35 per million output tokens.

This mix makes it a compelling candidate for cost-conscious coding assistance, structured SEO content, research, and tool-connected workflows. Its context supports longer source documents, while the agentic measurement offers some evidence for multi-step tasks. Teams should still test instruction precision and verify actions before assigning it unsupervised authority, particularly in client-facing workflows.

Consider it for routine-to-moderately complex workloads where low cost matters and outputs can be checked. Reserve a higher-capability model for the hardest reasoning tasks or work with little tolerance for error.

Assessed September 23, 2026

Editorial notes

Qwen3 235B A22B combines strong math and coding results, a 262K context, agentic capability, and low listed prices, making it a versatile value option.

The DFO score ranks models on the Artificial Analysis Intelligence Index and is refreshed as new models launch. The write-up and verdict are our own. Feedback?

DFO Verdict

Qwen3 235B A22B combines strong math and coding results, a 262K context, agentic capability, and low listed prices, making it a versatile value option.

#151 of 743 overall Up 73 this week

Benchmarks

Intelligence Index 12.5 #150 of 451 · Artificial Analysis
Coding Index 14.1 #132 of 199 · Artificial Analysis
Agentic Index 21.3 #41 of 181 · Artificial Analysis
Creative Writing Not rated Arena rating
Instruction Following Not rated Arena rating
GPQA Diamond 79%
HLE 15%
MMLU Pro 84.3%
MATH 500 98.4%
AIME 94%
AIME 2025 91%
SciCode 42.4%
LiveCodeBench 78.8%
TerminalBench Hard 13.6%
τ²-Bench 53.2%
IFBench 51.2%
LCR 67%

Artificial Analysis data refreshed Nov 11, 2026. Arena ratings from the Nov 9, 2026 text leaderboard, with style control. Bars: magenta = reasoning, ink = coding and agents, cyan = instructions and long context.

Artificial Analysis no longer lists this model, so its indexes are converted from their older scale to stay comparable.

How Qwen: Qwen3 235B A22B Instruct 2507 compares

Qwen: Qwen3 235B A22B Instruct 2507 ranks #150 of 451 AI models we track for overall intelligence, #132 of 199 for coding, #41 of 181 for agentic tasks. Its 262K-token context window is larger than 73% of the models we list. At $0.14 per million input tokens it is cheaper than 58% of comparable models.

Position in the field
DFO score: ahead of 80% of models #150
Speed: faster than 26% of models #262
Price: cheaper than 58% of models #312
Context: larger than 73% of models #202
worst in fieldmedianbest in field
Price vs frontier peers · $ per 1M tokens
Qwen: Qwen3 235B A22B Instruct 2507 $0.14 in $0.80 out
Anthropic: Claude Fable 5.2 $8.00 in $40.00 out
Google: Gemini 4 Carbon $2.00 in $10.00 out
Google: Gemini 4 Argon $2.00 in $10.00 out

Dark bar = input · light bar = output, scaled to the priciest peer.

Context window vs peers · tokens
Qwen: Qwen3 235B A22B Instruct 2507 262K

1M tokens ≈ 8 full-length novels or ~2,500 pages of business documents in a single request.

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.14 $0.000140
Output $0.80 $0.000800

What would Qwen: Qwen3 235B A22B Instruct 2507 cost your business?

Pick the job that looks most like yours, then fine-tune with the sliders. Estimates update live.

A website chatbot handling around 100 customer conversations a day, a few short messages each.

3,000
One request is one message, email, draft or automation call.
1,200 tokens

$1.22/mo Qwen: Qwen3 235B A22B Instruct 2507 $0.0004 per request
$63.36/mo Anthropic: Claude Fable 5.2 $0.02 per request
$15.84/mo Google: Gemini 4 Carbon · best value $0.0053 per request

Full calculator with 743 models → Price Calculator

DFO AI AUTOMATION

These numbers get smaller with the right architecture.

We route routine calls to cheap models and save Qwen: Qwen3 235B A22B Instruct 2507 for the hard ones. Most clients cut their estimate by 60-80%.

Talk to our team

About Qwen: Qwen3 235B A22B Instruct 2507

Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following,..

Frequently asked questions about Qwen: Qwen3 235B A22B Instruct 2507

How much does Qwen: Qwen3 235B A22B Instruct 2507 cost?

Qwen: Qwen3 235B A22B Instruct 2507 costs $0.14 per million input tokens and $0.80 per million output tokens.

What is the context window of Qwen: Qwen3 235B A22B Instruct 2507?

Qwen: Qwen3 235B A22B Instruct 2507 has a context window of 262,144 tokens (262K).

Is Qwen: Qwen3 235B A22B Instruct 2507 good for coding?

On our coding benchmark index, Qwen: Qwen3 235B A22B Instruct 2507 ranks #132 of 199 models, placing it in the broader range of the field for code generation and debugging.

What can Qwen: Qwen3 235B A22B Instruct 2507 do?

Qwen: Qwen3 235B A22B Instruct 2507 supports tool use and function calling.

Who created Qwen: Qwen3 235B A22B Instruct 2507?

Qwen: Qwen3 235B A22B Instruct 2507 is developed by Qwen and was released on July 21, 2025.

© 2026 Design for Online Ltd. Registered in England and Wales No. 10328553. VAT Registered. Design for Online® and Forerunner® are registered trademarks.