Qwen3 235B A22B

Name: Qwen3 235B A22B
Brand: qwen
Availability: OnlineOnly

Qwen:

3 235B A22B

A 235B mixture-of-experts model from Alibaba's Qwen team, activating 22B parameters per forward pass. Switches between a thinking mode for hard tasks and a non-thinking mode for chat. Strong on reasoning, tool calling and over 100 languages. 32K context, extendable to 131K.

Jailbreak1 of 9levels resisted

Website OpenRouter HuggingFace Docs GitHub Paper Blog

Provider: Qwen
Released: 2025-04-28
Parameters: 235B (22B Active)
Free API (OpenRouter): No
Rival’s tab: $0.36; 48/58 receipts survived

Free API availability

No free endpoint for Qwen3 235B A22B was found in Rival’s OpenRouter listings. Other providers or chat apps may have separate free offers.

Provider data: 13 Sep 2026OpenRouter usage limits

View API options

Benchmarks

Codeforces

Competitive with o3-mini/Gemini 2.5 Pro

source ↗

AIME

Competitive with o3-mini

source ↗

SWE-bench Verified

61.7%

source ↗

LiveCodeBench

70.7%

source ↗

Get API accessProvider and language code samples

Provider

fromimport openai  OpenAI

client = OpenAI(
"https://openrouter.ai/api/v1"    base_url=,
"$OPENROUTER_API_KEY"    api_key=,
)

response = client.chat.completions.create(
"qwen/qwen3-235b-a22b"    model=,
"role""user""content""Hello!"    messages=[{: , : }],
)
print(response.choices[0].message.content)

Set OPENROUTER_API_KEY with your OpenRouter API key from openrouter.ai/keys.

Deep analysisPsychometrics, taste index, writing DNA

SubjectiveBench

Taste Index

Across 58 scored outputs

How this is measured →

190.19xFloor

0100headroom →

Taste is judged on an uncapped scale, originality first. The space past 100 is craft today's models rarely reach.

Craft30

Originality14

Plays it safe

share of outputs that are the default answer

79%

Writing DNA

Stylometric Fingerprint

Based on 27 text responses

Tick = global average

Vocabulary Diversity59%

Unique words vs. total words. Higher = richer vocabulary.

Sentence Length16.2 words

Average words per sentence.

Hedging0.58

"Might", "perhaps", "arguably" per 100 words.

Bold Formatting7.0

**Bold** markers per 1,000 characters.

List Usage4.1

Bullet and numbered list items per 1,000 characters.

Section Structure0.88

Markdown headings per 1,000 characters.

Emoji Usage0.33

Emoji per 1,000 characters.

Transitions0.14

"However", "moreover", "furthermore" per 100 words.

Opening Habits

Consistency

76%

Across 27 responses

Favorites

Movie

The Godfather

19728.7

Album

Sgt Peppers Lonely Hearts Club Band

The Beatles

Book

Nineteen Eighty-Four

George Orwell

City

Istanbul

Turkey

Game

The Legend of Zelda: Ocarina of Time

Adventure, Action4.4

Model Responses

53 outputs · $0.35 tracked across 43 receipts

Reviewing agent-written code?See a Brief PR report

Keep exploring

COMPARE

Qwen3 235B A22B vs Ox Alpha

Real outputs compared side by side

RANKINGS

Best AI for Creative Writing

A stand-up set with puns banned, a manifesto against CAPTCHA, an interview wi...

Compare Qwen3 235B A22B

Alternatives to Qwen3 235B A22B

Qwen3 235B A22B

Free API availability

Benchmarks

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

Qwen3 235B A22B vs Ox Alpha

Best AI for Creative Writing

Compare Qwen3 235B A22B

Alternatives to Qwen3 235B A22B

Qwen3 235B A22B

Free API availability

Benchmarks

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

Qwen3 235B A22B vs Ox Alpha

Best AI for Creative Writing