The LLM benchmark aggregator.

Qwen3.5 2B

Alibaba · 2B

Strongest in Knowledge work (#150 of 177, on 1 of its 2 boards), weakest in Agents (#232 of 267). Above par in 3 of 8 scopes. Among the models it meets almost everywhere, it finishes behind Claude Sonnet 5 and GLM-5.2 and ahead of Kimi K2.6 and GPT-5.4 Mini.

§ 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.

Agents41.1−27#232/2671/5

Knowledge work36.7−36.9#150/1771/2

Tool use52−22—/81✱0/1

Reasoning49.2−17.7—/178✱0/4

Mathematics49.3−16.4—/140✱0/2

Science43.4−20.1—/122✱0/1

Coding51.4−17.7—/164✱0/5

Code generation53.6−16.4—/77✱0/2

§ 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 Qwen3.5 2B, left for the other.

GPT-4owon 5 · lost 0

Claude 3.5 Sonnetwon 5 · lost 0

Llama 4 Scout Instructwon 6 · lost 0

DeepSeek V3won 7 · lost 0

Qwen3.5 0.8Bwon 8 · lost 0

NVIDIA Nemotron 3 Nano 4Bwon 8 · lost 0

GPT-4.1 Nanowon 6 · lost 1

Mistral Smallwon 4 · lost 1

Mistral Largewon 4 · lost 1

Granite 4.2 3Bwon 6 · lost 2

Nemotron 3 Ultrawon 5 · lost 2

GPT-4o Miniwon 5 · lost 2

Llama 4 Maverick Instructwon 5 · lost 2

OpenBMB 1Bwon 5 · lost 2

G9v3 3Bwon 5 · lost 2

Gemma 4 E4Bwon 4 · lost 2

Claude 3.5 Haikuwon 4 · lost 2

Nemotron 3 Superwon 4 · lost 2

Mistral Medium 3.5won 4 · lost 2

G9v3 39A5Bwon 4 · lost 2

Qwen3.6 27Bwon 4 · lost 2

Nemotron 3 Ultra 550B A55Bwon 4 · lost 2

Kimi K2.7 Codewon 4 · lost 2

Command Awon 4 · lost 2

Grok Build 0.1won 4 · lost 2

K2 Horizon 0.9Bwon 5 · lost 3

Inklingwon 5 · lost 3

MiniMax M3won 5 · lost 3

Claude Opus 4won 3 · lost 2

Mistral Small 3.1won 3 · lost 2

Mercury 2.5won 3 · lost 2

Gemini 2.5 Flash Litewon 3 · lost 2

Granite 4.2 30Bwon 4 · lost 3

Qwen3.6 35B A3Bwon 4 · lost 3

Gemini 3.5 Flash Litewon 4 · lost 3

Grok 4.3won 4 · lost 3

GPT-5.4 Miniwon 4 · lost 3

Kimi K2.6won 4 · lost 3

Mistral Large 3won 3 · lost 3

K2 Horizon 32Bwon 3 · lost 3

Inkling Smallwon 3 · lost 3

Mimo V2.5won 3 · lost 3

Mimo V2.5 Prowon 3 · lost 3

GPT-4.1won 3 · lost 3

DeepSeek V4 Flash Visionwon 3 · lost 3

Qwen3.8 Flash Nextwon 3 · lost 3

O3 Miniwon 3 · lost 4

Granite 4.2 8Bwon 3 · lost 4

Qwen3.5 9Bwon 3 · lost 4

GPT-4.1 Miniwon 3 · lost 4

Qwen3.5 4Bwon 3 · lost 4

Gemini 3.5 Flashwon 3 · lost 4

Qwen3.6 Pluswon 3 · lost 4

Qwen3.7 Maxwon 3 · lost 4

Grok 4.5won 3 · lost 4

Gemini 3.6 Flashwon 3 · lost 4

Gemini 2.5 Flashwon 3 · lost 4

GLM-5.3 Flashwon 3 · lost 4

Claude Opus 4.6won 2 · lost 3

Magistral Small 1.2won 2 · lost 3

Magistral Medium 1.2won 2 · lost 3

Claude 3.7 Sonnetwon 2 · lost 3

GPT-5 Nanowon 2 · lost 3

Qwen3.8 27Bwon 3 · lost 5

DeepSeek R1won 2 · lost 4

Gemma 4 12Bwon 2 · lost 4

GPT OSS 120Bwon 2 · lost 4

Kimi K2won 2 · lost 4

Gemma 4 31Bwon 2 · lost 5

Gemini 3.1 Prowon 2 · lost 5

Claude Sonnet 4won 2 · lost 5

GPT-5.4 Nanowon 2 · lost 5

K2 Horizon 7Bwon 2 · lost 5

Claude Haiku 4.5won 2 · lost 5

K2 Horizon MoVA 36B A4Bwon 2 · lost 5

GPT-5.2won 2 · lost 5

Muse Spark 1.1won 2 · lost 5

DeepSeek V4 Flashwon 2 · lost 5

GLM-5.3won 2 · lost 5

K2 Horizon 3.7Bwon 2 · lost 6

Ling 3.0 Flashwon 1 · lost 4

Gemini 3.1 Flash Litewon 1 · lost 5

Muse Sparkwon 1 · lost 5

MiniMax M2.7won 1 · lost 5

GPT-5.1won 1 · lost 5

Muse Glimmer-30Bwon 1 · lost 5

GLM-4.7won 1 · lost 5

GLM-5.1won 1 · lost 5

Kimi K2.5won 1 · lost 5

GPT-6 Lunawon 1 · lost 5

Muse Spark 1.2won 1 · lost 5

Grok 4.1 Fastwon 1 · lost 5

Gemini 3 Prowon 1 · lost 5

Grok 4.7won 1 · lost 5

Gemini 2.5 Prowon 1 · lost 6

GPT-5 Miniwon 1 · lost 6

DeepSeek V3.2won 1 · lost 6

Qwen3 235B A22B Instructwon 1 · lost 6

GPT-5.4won 1 · lost 6

O4 Miniwon 1 · lost 6

Gemini 3.7 Flashwon 1 · lost 6

Claude Sonnet 4.5won 1 · lost 6

Claude Sonnet 4.6won 1 · lost 6

Gemini 3.8 Flashwon 1 · lost 6

GLM-4.6won 1 · lost 6

DeepSeek V4 Prowon 1 · lost 6

Claude Opus 4.5won 1 · lost 6

Grok 4.6won 1 · lost 6

Qwen3.8 Maxwon 1 · lost 6

GPT-5.6 Terrawon 1 · lost 7

GPT-5.6 Lunawon 1 · lost 7

Grok 3 Miniwon 0 · lost 5

GPT OSS 20Bwon 0 · lost 5

Mimo V2.6 Prowon 0 · lost 5

GPT-5won 0 · lost 6

Grok 4won 0 · lost 6

GPT-6 Solwon 0 · lost 6

GPT-6 Astrawon 0 · lost 6

Muse Spark 1.3won 0 · lost 6

Claude Opus 5.5won 0 · lost 6

O3won 0 · lost 7

GPT-5.5won 0 · lost 7

Claude Opus 4.7won 0 · lost 7

K2 Horizon 375B A23Bwon 0 · lost 7

Claude Opus 4.8won 0 · lost 7

DeepSeek V4.1 Flashwon 0 · lost 7

GPT-5.6 Solwon 0 · lost 7

Kimi K3won 0 · lost 7

Claude Fable 5.1won 0 · lost 7

Claude Fable 5won 0 · lost 7

Claude Opus 5won 0 · lost 7

GLM-5.2won 0 · lost 8

Claude Sonnet 5won 0 · lost 8

§ 3 · Sources

Where the numbers come from

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

GDPval-AA ↗1 measureread 2026-09-27

GDPval-AA -61

✱✱ IFM — K2 Horizon launch ↗the model’s own publisher8 measuresread 2026-09-27

GPQA Diamond 54.9%HMMT Feb 2026 22.7%BFCL v4 43.6%HumanEval+ 75.6%MBPP+ 67.7%AIME 2025 34.2%AIME 2026 38.8%LiveCodeBench v6 29.8%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=qwen3-5-2b)](https://publicai.io/model-index/m/qwen3-5-2b)