The LLM benchmark aggregator.

Qwen3 30B A3B Thinking

Alibaba · 30B · A3B

Strongest in General intelligence (#79 of 181, on 1 of its 3 boards), weakest in Agents (#297 of 298). Above par in 2 of 4 scopes. Among the models it meets almost everywhere, it finishes behind Gemini 4 Carbon and Claude Fable 5 and ahead of GPT-4.1 and Kimi K2.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Core abilities50.7−14.9#90/2011/4

General intelligence51.1−17.9#79/1811/3

Agents38.3−28#297/2982/6

Knowledge work31.3−39.4#200/2082/2

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Qwen3 30B A3B Thinking, left for the other.

Gemma 4 31Bwon 3 · lost 1

Gemma 3 27B ITwon 3 · lost 1

UI-TARS 1.5 7Bwon 2 · lost 1

Smaug Nemotron 3.5won 2 · lost 1

Gemma 3 4B ITwon 2 · lost 1

Command Awon 2 · lost 1

Qwen2.5 Vl 72B Instructwon 2 · lost 1

Gemini 2.5 Flash-Litewon 2 · lost 1

Kimi K2won 2 · lost 1

GPT-4.1won 2 · lost 1

NVIDIA Nemotron 3.5 Lightning 30B A3Bwon 2 · lost 2

Nemotron 3.5 Super VLwon 2 · lost 2

GPT-5.4 miniwon 2 · lost 2

Muse Spark Openwon 2 · lost 2

Nemotron 3 Nano 4Bwon 2 · lost 2

Mistral Small 4won 2 · lost 2

GPT-5.4 nanowon 2 · lost 2

North Mini Codewon 2 · lost 2

GPT-4o miniwon 2 · lost 2

Mistral Medium 3.5won 2 · lost 2

Granite 4.2 3Bwon 2 · lost 2

Granite 4.2 8Bwon 2 · lost 2

Command A+won 2 · lost 2

Muse Glimmerwon 2 · lost 2

MiniMax M3won 2 · lost 2

GPT-5.6 Lunawon 2 · lost 2

GPT-4.1 nanowon 2 · lost 2

K-EXAONE 2.0won 2 · lost 2

Mistral Large 4won 2 · lost 2

GPT-6 Lunawon 2 · lost 2

Inklingwon 2 · lost 2

Step 5won 2 · lost 2

Gemini 3.5 Flash-Litewon 2 · lost 2

Llama 4 Scout Instructwon 2 · lost 2

GLM-5.3 Flashwon 2 · lost 2

Inkling Smallwon 2 · lost 2

Apodex 1.1won 2 · lost 2

Solar Pro 4won 2 · lost 2

GPT OSS 120Bwon 2 · lost 2

Mistral Small 3.2won 2 · lost 2

Beam 1.0won 2 · lost 2

K2 Horizon 375B A23Bwon 2 · lost 2

NVIDIA Nemotron 3 Ultra 550B A55Bwon 2 · lost 2

JT-4.1 Flash 236B A21Bwon 2 · lost 2

GLM-4.6won 2 · lost 2

Nanbeige4.3 33Bwon 2 · lost 2

GPT-4.1 miniwon 2 · lost 2

Qwen3 14Bwon 2 · lost 2

Qwen3.8 27Bwon 2 · lost 2

DeepSeek V4 Flashwon 2 · lost 2

MiniMax M3.1 Flashwon 2 · lost 2

Agnes 2.5 Pro Thetawon 2 · lost 2

Mistral Large 3won 2 · lost 2

DeepSeek V3.2won 2 · lost 2

DeepSeek V3won 2 · lost 2

GPT OSS 20Bwon 2 · lost 2

MiMo V2.6 Flashwon 2 · lost 2

DeepSeek V4 Prowon 2 · lost 2

Qwen3.6 27Bwon 1 · lost 2

Qwen3.6 Pluswon 1 · lost 2

Kimi K2.7 Codewon 1 · lost 2

Grok 4.3won 1 · lost 2

Grok Build 0.1won 1 · lost 2

Claude Sonnet 5won 1 · lost 2

Claude Sonnet 4.6won 1 · lost 2

Kimi K2.6won 1 · lost 2

Muse Spark 1.1won 1 · lost 2

Claude Haiku 5.5won 1 · lost 3

GLM-5.2won 1 · lost 3

MiMo V2.6 Prowon 1 · lost 3

GPT-5.6 Terrawon 1 · lost 3

Gemini 3.8 Flashwon 1 · lost 3

Grok 4.6won 1 · lost 3

Qwen3.8 Maxwon 1 · lost 3

Kimi K3won 1 · lost 3

Qwen3 235B A22B Instructwon 0 · lost 3

Gemini 2.5 Flashwon 0 · lost 3

GPT-5.2won 0 · lost 3

Claude Opus 5won 0 · lost 3

DeepSeek V4 Flash Visionwon 0 · lost 3

GPT-5 nanowon 0 · lost 3

Claude Opus 4.5won 0 · lost 3

GPT-6 Solwon 0 · lost 3

Muse Spark 1.2won 0 · lost 3

Qwen3.7 Maxwon 0 · lost 3

O4 Miniwon 0 · lost 3

O3won 0 · lost 3

Grok 4won 0 · lost 3

Gemini 3.6 Flashwon 0 · lost 3

Gemini 3.5 Flashwon 0 · lost 3

Gemini 3 Prowon 0 · lost 3

Gemini 3.1 Prowon 0 · lost 3

Gemini 3.7 Flashwon 0 · lost 3

Llama 4 Maverick Instructwon 0 · lost 4

Ling 3.1 Flashwon 0 · lost 4

DeepSeek V3.1 Terminuswon 0 · lost 4

Claude Sonnet 4.5won 0 · lost 4

Inkling 2 Smallwon 0 · lost 4

GLM-5.3won 0 · lost 4

Tencent Hy4won 0 · lost 4

Muse Emberwon 0 · lost 4

Qwen3.8 Flash Nextwon 0 · lost 4

Claude Sonnet 5.5won 0 · lost 4

Claude Opus 4.7won 0 · lost 4

DeepSeek V4.1 Flashwon 0 · lost 4

Qwen3 Next 80B A3B Thinkingwon 0 · lost 4

GPT-5.4won 0 · lost 4

Gemini 3.1 Flash-Litewon 0 · lost 4

Grok 4.7won 0 · lost 4

Qwen3 235B A22B Thinkingwon 0 · lost 4

DeepSeek R1won 0 · lost 4

Gemini 2.5 Prowon 0 · lost 4

GPT-5 miniwon 0 · lost 4

Kimi K3.1 Prowon 0 · lost 4

GPT-5won 0 · lost 4

Claude Sonnet 4won 0 · lost 4

Qwen3.5 397B A17Bwon 0 · lost 4

Muse Spark 1.3won 0 · lost 4

GPT-5.6 Solwon 0 · lost 4

Muse Wonderwon 0 · lost 4

Muse Wonder Fullwon 0 · lost 4

Kimi K2.5won 0 · lost 4

Grok 4.5won 0 · lost 4

Claude Opus 4.8won 0 · lost 4

Claude Opus 5.5won 0 · lost 4

Claude Fable 5.1won 0 · lost 4

GPT-6.1 Terrawon 0 · lost 4

GPT-5.5won 0 · lost 4

Gemini 4 Argonwon 0 · lost 4

GPT-6.1 Solwon 0 · lost 4

Claude Fable 5.2won 0 · lost 4

GPT-6 Astrawon 0 · lost 4

Claude Fable 5won 0 · lost 4

Gemini 4 Carbonwon 0 · lost 4

§ 3 · Sources

Where the numbers come from

3 publications, 3 figures. Every one links to the page it was read from.

GDPval-AA ↗1 measureread 2026-11-12

GDPval-AA 75

AA-Briefcase ↗1 measureread 2026-11-12

AA-Briefcase 6

Kagi LLM Benchmark ↗1 measureread 2026-11-12

Kagi LLM Benchmark 54.9%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=qwen3-30b-a3b-thinking)](https://publicai.io/model-index/m/qwen3-30b-a3b-thinking)