Qwen3-235B-A22B

Name: Qwen3-235B-A22B
Author: Alibaba

Released

2025-04-29

Last refreshed

2026-10-26

Status

Researched 19d ago

Open sourceCommercial use: permittedCodingRAGLong contextClassificationJSON / Tool use

Alibaba releases · 51 in the last 12 months · this family litChangelog →

Specifications

Family: Qwen3
Released: 2025-04-29
Context: 128K
Parameters: 235B
Architecture: Decoder Only
Specialization: general
Openness: Open source
License: Apache 2.0OSI-approvedPermitted
Weights: Available
Code: Unknown
Training: Pretrained

Created by

Alibaba

AI research institute of Alibaba Group.

Hangzhou, Zhejiang, China

Founded 2017

Website

Pricing

No tracked provider pricing is available yet.

Providers(2)

Fireworks AI Venice AI

View 2 provider routes

Created by

Alibaba

AI research institute of Alibaba Group.

Hangzhou, Zhejiang, China

Founded 2017

Website

Pricing

No tracked provider pricing is available yet.

Providers(2)

Fireworks AI Venice AI

View 2 provider routes

About

Qwen3-235B-A22B is Alibaba's Qwen3 model. It offers a 128K-token context window.

Top use-case fit: coding, agents, and build tasks

Coding

Q/$ C

RAG

Q/$ D

Long context

Q/$ D

Providers(2)

Compare all 2

Compare API pricing across 2 tracked provider routes for input and output tokens, batch, and cached reads when available.

Provider	Input / 1M	Output / 1M	Route
Fireworks AI	-	-	ProvisionedPartial
Venice AI	-	-	ServerlessPartial

Deprecated provider routes(1)

Provider	Status	Catalog ID
OpenRouter	DeprecatedRetired October 19, 2026	qwen/qwen3-235b-a22b

Available via routers & gateways(1)

OpenRouter

Hybrid

Unified hybrid gateway to 500+ models from 80+ providers via a single OpenAI-compatible API, with an optional Auto Router that selects a model per prompt from market spend data.

Passthrough + feeFireworks AI

Capabilities

Structured Outputs

Benchmark peer barsfor Coding

HumanEvalRank 10 of 99

Claude Sonnet 4.6

98.0

96.7

Claude Opus 4.6

95.0

Grok-3

94.5

Qwen3-235B-A22Bcurrent

92.7

LiveCodeBenchRank 34 of 79

DeepSeek V4 Pro

93.5

Qwen3.8-Omni-Flash

92.6

Qwen3.8-Flash

91.9

Gemini 3.1 Pro Preview

91.7

Qwen3-235B-A22Bcurrent

80.4

Aider PolyglotRank 18 of 34

88.0

88.0

84.9

83.1

Qwen3-235B-A22Bcurrent

59.6

Benchmark scores(9)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.

Benchmark	Score	Version	Evaluation	Source
HumanEval	92.7	2025-04Observed 2026-04-09	—	Source
LiveCodeBench	80.4	2026-04Observed 2026-04-14	—	Source
Aider Polyglot	59.6	2026-04Observed 2026-04-14	—	Source
Google-Proof Q&A	86.1	diamondObserved 2026-04-18	—	Source
MMLU PRO	82.8	—Observed 2025-04-29	—	Source
AIME 2024	85.7	—Observed 2025-04-29	—	Source
RULER	90.6	RULER (128K) (accuracy)Observed 2026-10-26	—	Source
Chatbot Arena	1402.0	text_style_control overall 2026-10-22 (style control, default view)Observed 2026-10-26	Configuration: Qwen3-235B-A22B (Non-Thinking) Harness: LMArena blind pairwise human votes, Bradley-Terry Evaluator: LMArena Confidence: confirmed Notes: LMArena text style control leaderboard, overall category, published 2026-10-22 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Qwen3-235B-A22B (Non-Thinking)': rank 161, Arena Score 1402.1 (95% CI 1397.6-1406.6), 37485 votes. Retrieved 2026-10-26.	Source Source: LMArena (lmarena.ai), CC BY 4.0
LMArena Text Arena - Creative Writing	1356.0	text creative_writing 2026-10-22Observed 2026-10-22	Configuration: Qwen3-235B-A22B (Non-Thinking) Harness: LMArena blind pairwise human votes, Bradley-Terry Evaluator: LMArena Confidence: confirmed Notes: LMArena Text arena, Creative Writing category (no style control), board published 2026-10-22 (HF lmarena-ai/leaderboard-dataset @a467e6b9, text config, latest split). Board entry 'Qwen3-235B-A22B (Non-Thinking)' (category creative_writing in text/latest-00000-of-00001.parquet): rank 161 of 416, rating 1356.2 (95% CI 1347.2-1365.3), 4569 votes. Other board entries for this model, not used: 'Qwen3-235B-A22B' 1316.1 (rank 209). Source: LMArena (CC BY 4.0). Retrieved 2026-10-27.	Source Source: LMArena (lmarena.ai), CC BY 4.0

Migration checks

No linked migration route is available for this model yet.

Rankings & picks(1)

Best LLMs for RAGListed

Compare Qwen3-235B-A22B with other models

Comparison and alternatives

Browse all comparisons →

Qwen3-235B-A22B vs Step 3.7 Flash Qwen3-235B-A22B vs DeepSeek R1 Qwen3-235B-A22B vs Claude Opus 4.6 Qwen3-235B-A22B vs Claude Sonnet 4.6 Qwen3-235B-A22B vs DeepSeek V4 Flash Qwen3-235B-A22B vs Grok 4 Qwen3-235B-A22B vs Kimi K2.5 Qwen3-235B-A22B vs DeepSeek V4 Pro Qwen3-235B-A22B vs DeepSeek V3 Qwen3-235B-A22B vs Claude Sonnet 4.5 Qwen3-235B-A22B vs Qwen3.6-27B Qwen3-235B-A22B vs Claude Opus 4.7 Qwen3-235B-A22B vs Grok 4.3 Qwen3-235B-A22B vs Claude Opus 4.5 Qwen3-235B-A22B vs GPT-5.4 Qwen3-235B-A22B vs Grok 4.20 Multi-Agent

Show all 30 popular comparisonssorted by 7-day search impressions