Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. Models
  2. Text
  3. Qwen

Qwen 3 235b A22B 2507 Thinking

The thinking version of Qwen 3 235b A22B 2507, with enhanced reasoning capabilities and step-by-step problem solving.

Added Sep 11, 2025

Model weights

Pricing

Auto routing · per 1M tokens
Input
$0.30
Output
$0.50
Compare provider prices

Specifications

Context window
256K
Max output
262.1K
Parameters
235B / 22B
Total / active
Avg output (7d)
2.2K tokens
Longer than 94% of models

Benchmarks

Sourced from LMArena.

Arena Score

1400.3

Overall Rank

#168 / 421

Votes

8,812

Confidence Interval

1393.7 - 1406.8

Category Scores

Coding

#172 / 416

1,590 votes

1442.0

Math

#168 / 404

476 votes

1397.0

Longer Query

#175 / 399

1,637 votes

1402.3

Creative Writing

#161 / 419

1,083 votes

1369.8

Instruction Following

#171 / 421

2,093 votes

1385.4

Hard Prompts

#173 / 421

3,804 votes

1417.8

Additional Categories
21

Expert

#96 / 371

423 votes

1468.0

Japanese

#115 / 288

442 votes

1376.3

Industry Medicine And Healthcare

#130 / 390

507 votes

1440.6

Korean

#130 / 289

204 votes

1362.2

German

#134 / 320

211 votes

1400.9

Chinese

#135 / 395

535 votes

1454.3

Industry Mathematical

#138 / 401

475 votes

1421.7

Russian

#145 / 385

527 votes

1405.3

Polish

#158 / 244

784 votes

1379.1

Industry Business And Management And Financial Operations

#159 / 414

1,525 votes

1404.7

Non English

#161 / 421

4,911 votes

1389.8

Hard Prompts English

#166 / 419

1,715 votes

1427.4

Exclude Ties

#168 / 421

6,201 votes

1382.1

Industry Software And It Services

#169 / 421

2,907 votes

1433.8

Industry Writing And Literature And Language

#171 / 420

1,859 votes

1377.1

Industry Life And Physical And Social Science

#173 / 419

1,489 votes

1412.8

Spanish

#173 / 303

185 votes

1381.6

Industry Legal And Government

#177 / 393

532 votes

1397.2

Multi Turn

#180 / 419

1,355 votes

1392.7

English

#181 / 421

3,893 votes

1405.6

Industry Entertainment And Sports And Media

#183 / 419

1,526 votes

1350.4

Published 2026-11-09 · Matched as Qwen3-235B-A22B-Thinking (2507)

LMArena Dataset

Providers

Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)

Auto
Auto (ZDR)
$0.45$3.50N/AN/AN/A
Auto
Auto (Anonymized)
$0.30$0.50N/AN/AN/A
Venice
Venice
FP8
US
$0.472$3.68N/A0.9s34.5 tps
Privacy:ZDR, prompts not retainedNo training on promptsBest latency and throughput are highlighted.
Batch API
ProviderInputOutputCached input
Auto
Autoqwen/qwen3-235b-a22b-thinking-2507
$0.30/M$0.50/MN/A

Batch rows for this model use normal routing and cost the same as standard API requests. Set provider in a row to choose a provider.

Related text models

Compare Qwen 3 235b A22B 2507 Thinking with similar models from the same provider or model family.

Qwen 3 235b A22B 2507

qwen/qwen3-235b-a22b-instruct-2507

Qwen 3 235b A22B Instruct 2507 the updated version of Qwen3 235B A22B, with significant improvements in performance. This model is non-thinking.

Qwen 3 235b A22B

qwen/qwen3-235b-a22b

Qwen 3 235b is a 235b model with 22B active parameters. Supports switching between thinking and non thinking: trigger thinking with /think and /no_think anywhere in a prompt or system message to toggle chain-of-thought reasoning.

Qwen3 VL 235B A22B Instruct

qwen/qwen3-vl-235b-a22b-instruct

Routed across multiple providers, which may include Alibaba, a Chinese entity; provider-specific privacy and logging guarantees may vary. Qwen3 Vision‑Language model (235B MoE, ≈22B active) tuned for instruction following and grounded visual QA. Excels at image understanding, dense OCR, charts/diagrams, and multi‑image context.

Qwen 3.8 27B Antislop

qwen/qwen3.8-27b-antislop

Qwen 3.8 27B Antislop is a creative-writing finetune designed to reduce formulaic prose. Available capabilities include image understanding, tool calling, and optional reasoning, with context windows up to 262,144 tokens.

Qwen 3.8 27B DarkIdol

qwen/qwen3.8-27b-darkidol

Qwen 3.8 27B DarkIdol is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.

Qwen 3.8 27B Fimi

qwen/qwen3.8-27b-fimi

Qwen 3.8 27B Fimi is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.