The LLM benchmark aggregator.

Qwen3 Next 80B A3B Thinking

Alibaba · 81B · A3B · open weights

Strongest in Safe-prompt compliance (#9 of 82), weakest in Agents (#210 of 268). Above par in 9 of 11 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and Grok 4.7 and ahead of Gemini 3.6 Flash and Gemini 3.5 Flash.

§ 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 abilities54.8−12.6#49/2041/3

General intelligence56.9−13.2#47/2041/3

Safety54.5−7#59/3372/3

Safe-prompt compliance59.3−1.1#9/821/1

Factual grounding51.7−18.9#46/1011/1

Harm refusal55.4−6.2#65/3001/2

Fairness52.5−17.5#91/3001/2

Human preference54−13.5#149/3421/1

Agents43.1−25#210/2681/5

Knowledge work39.7−33.9#134/1781/2

§ 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 Next 80B A3B Thinking, left for the other.

Llama 3.1 8B Instructwon 7 · lost 0

GPT-4won 7 · lost 0

Magistral Mediumwon 7 · lost 0

Llama 3 70B Instructwon 8 · lost 0

Gemini 2.0 Flashwon 8 · lost 0

GPT-4o Miniwon 10 · lost 0

GPT-4.1 Nanowon 10 · lost 0

Llama 4 Scout Instructwon 10 · lost 1

Llama 4 Maverick Instructwon 10 · lost 1

Gemma 3 27B Itwon 9 · lost 1

Gemma 3 4B Itwon 8 · lost 1

Command Awon 8 · lost 1

Claude 3 Haikuwon 7 · lost 1

Gemma 3 12B Itwon 7 · lost 1

Trinity Largewon 7 · lost 1

Command R Pluswon 6 · lost 1

Mistral Smallwon 7 · lost 2

Qwen3 32Bwon 6 · lost 2

O3 Miniwon 6 · lost 2

Mistral Large 3won 6 · lost 2

DeepSeek V3won 8 · lost 3

Kimi K2.7 Codewon 5 · lost 2

Nemotron 3 Ultra 550B A55Bwon 5 · lost 2

Grok Build 0.1won 5 · lost 2

Ling 3.0 Flash VLwon 5 · lost 2

Nova Microwon 5 · lost 2

GPT-4owon 5 · lost 2

MiniMax M2won 5 · lost 2

NVIDIA Nemotron 3 Super 120B A12Bwon 5 · lost 2

GPT OSS 120Bwon 7 · lost 4

Grok 3won 5 · lost 3

GPT OSS 20Bwon 6 · lost 4

GPT-5 Nanowon 6 · lost 4

GPT-4.1 Miniwon 6 · lost 4

GLM-4.6won 6 · lost 4

Nova Prowon 4 · lost 3

Mistral Largewon 4 · lost 3

Mistral Mediumwon 4 · lost 3

GLM-4.5won 4 · lost 3

Grok 4.1 Fastwon 4 · lost 3

GLM-4.5 Airwon 5 · lost 4

DeepSeek V4 Flashwon 5 · lost 4

MiniMax M3won 5 · lost 4

Grok 4.3won 5 · lost 4

GPT-5.6 Lunawon 5 · lost 4

Qwen3.7 Maxwon 5 · lost 4

GLM-5.2won 5 · lost 4

Gemini 3.5 Flashwon 5 · lost 4

Gemini 3.6 Flashwon 5 · lost 4

Claude Opus 4.6won 4 · lost 4

Gemini 2.5 Flash Litewon 5 · lost 5

Qwen3.5 122B A10Bwon 5 · lost 5

Kimi K2won 5 · lost 5

DeepSeek V3.2won 5 · lost 5

Gemma 4 31Bwon 5 · lost 5

Kimi K2.6won 5 · lost 5

Claude Sonnet 4.6won 5 · lost 5

DeepSeek R1won 5 · lost 6

Grok 4won 4 · lost 5

GPT-6 Lunawon 4 · lost 5

Mimo V2.5 Prowon 4 · lost 5

GLM-5.3 Flashwon 4 · lost 5

Phi 4won 3 · lost 4

GLM-5.1won 3 · lost 4

Claude Opus 4.5won 3 · lost 4

O4 Miniwon 4 · lost 6

Gemini 2.5 Flashwon 4 · lost 6

GPT-4.1won 4 · lost 6

Kimi K2.5won 4 · lost 6

DeepSeek V4 Prowon 4 · lost 6

Gemini 3 Prowon 4 · lost 6

Qwen3.5 35B A3Bwon 3 · lost 5

Gemini 3.1 Flash Litewon 3 · lost 5

Gemma 4 26B A4Bwon 3 · lost 5

GPT-5.4 Miniwon 3 · lost 5

Gemini 3.1 Prowon 3 · lost 5

Claude Haiku 4.5won 4 · lost 7

Claude Sonnet 4.5won 4 · lost 7

GPT-5.2won 3 · lost 6

Qwen3.5 397B A17Bwon 3 · lost 6

Qwen3.8 Maxwon 3 · lost 6

Mimo V2.6 Prowon 3 · lost 6

Qwen3 235B A22B Instructwon 3 · lost 7

GPT-5.6 Solwon 3 · lost 7

Claude 3.5 Haikuwon 2 · lost 5

Claude 3.7 Sonnetwon 2 · lost 5

Claude Sonnet 4won 3 · lost 8

Gemini 2.5 Prowon 3 · lost 8

GPT-5.4 Nanowon 2 · lost 6

Claude Opus 4won 2 · lost 7

Inklingwon 2 · lost 7

GPT-5.1won 2 · lost 7

Hy3won 2 · lost 7

Claude Sonnet 5won 2 · lost 7

GPT-5.6 Terrawon 2 · lost 7

Muse Spark 1.1won 2 · lost 7

Gemini 3.8 Flashwon 2 · lost 7

GPT-6 Solwon 2 · lost 8

GPT-5.4won 2 · lost 8

Claude Opus 4.7won 2 · lost 8

GPT-5 Miniwon 2 · lost 9

GPT-5won 2 · lost 9

GPT-6 Astrawon 1 · lost 7

O3won 1 · lost 8

Grok 4.6won 1 · lost 8

Grok 4.5won 1 · lost 8

Claude Opus 4.8won 1 · lost 8

Claude Opus 5won 1 · lost 8

GPT-5.5won 1 · lost 9

MiniMax M2.7won 0 · lost 7

Grok 4.7won 0 · lost 9

Claude Opus 5.5won 0 · lost 9

§ 3 · Sources

Where the numbers come from

5 publications, 9 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26

LMArena Text 1369

GDPval-AA ↗1 measureread 2026-09-26

GDPval-AA 114

Kagi LLM Benchmark ↗1 measureread 2026-09-26

Kagi LLM Benchmark 66.7%

HELM Safety ↗5 measuresread 2026-09-26

HELM Safety · HarmBench 84.6%HELM Safety · SimpleSafetyTests 99.5%HELM Safety · Anthropic Red Team 99.9%HELM Safety · BBQ 91.7%HELM Safety · XSTest 98.3%

Vectara Hallucination Leaderboard ↗1 measureread 2026-09-26

Vectara · Factual consistency 90.7%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=qwen3-next-80b-a3b-thinking)](https://publicai.io/model-index/m/qwen3-next-80b-a3b-thinking)