The LLM benchmark aggregator.

Qwen3 14B

Alibaba · 15B · open weights

Strongest in Factual grounding (#17 of 104), weakest in Agents (#211 of 298). Above par in 3 of 7 scopes. Among the models it meets almost everywhere, it finishes behind DeepSeek V4.1 Flash and Claude Opus 4.8 and ahead of GLM-4.6 and NVIDIA Nemotron 3 Ultra 550B A55B.

§ 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.

Safety53.6−7.8#80/3271/3

Factual grounding61.5−9.2#17/1041/1

Core abilities48.8−16.8#114/2001/4

General intelligence48.2−20.8#107/1801/3

Agents45.5−20.8#211/2982/6

Tool use52.3−21.7#33/811/1

Knowledge work36−34.7#177/2081/2

§ 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 14B, left for the other.

Mistral Small 3.2won 5 · lost 0

Gemma 3 4B ITwon 6 · lost 0

GPT-4.1 nanowon 6 · lost 0

Llama 4 Scout Instructwon 7 · lost 0

Gemma 3 27B ITwon 7 · lost 0

Gemma 4 31Bwon 5 · lost 1

Gemma 3 12B ITwon 4 · lost 1

Llama 3.3 70B Instructwon 4 · lost 1

GPT-4o miniwon 4 · lost 1

Llama 4 Maverick Instructwon 5 · lost 2

GPT-5.4 miniwon 4 · lost 2

Command Awon 4 · lost 2

DeepSeek V3won 4 · lost 2

DeepSeek R1won 4 · lost 2

MiniMax M3won 3 · lost 2

Mistral Large 4won 3 · lost 2

Inklingwon 3 · lost 2

Claude Sonnet 4.6won 3 · lost 2

Kimi K2.6won 3 · lost 2

NVIDIA Nemotron 3 Ultra 550B A55Bwon 3 · lost 2

GLM-4.6won 4 · lost 3

GPT OSS 120Bwon 3 · lost 3

Gemini 2.5 Flash-Litewon 3 · lost 3

GPT-4.1 miniwon 3 · lost 3

Mistral Large 3won 3 · lost 3

Gemini 3.1 Flash-Litewon 3 · lost 3

Gemini 2.5 Prowon 3 · lost 3

Amazon Nova 2 Litewon 2 · lost 3

Qwen3 8Bwon 2 · lost 3

GPT-6 Lunawon 2 · lost 3

GLM-5.3 Flashwon 2 · lost 3

GLM-5.2won 2 · lost 3

GPT-6 Solwon 2 · lost 3

Gemini 3.1 Prowon 2 · lost 3

GPT-5.4 nanowon 2 · lost 4

Kimi K2won 2 · lost 4

DeepSeek V4 Prowon 2 · lost 4

Gemini 2.5 Flashwon 2 · lost 4

Claude Opus 4.5won 2 · lost 4

Qwen3 Next 80B A3B Thinkingwon 2 · lost 4

O4 Miniwon 2 · lost 4

GPT-5.6 Solwon 2 · lost 4

Kimi K2.5won 2 · lost 4

Gemini 3 Prowon 2 · lost 4

Gemini 4 Argonwon 2 · lost 4

GPT-6 Astrawon 2 · lost 4

DeepSeek V3.2won 2 · lost 5

Qwen3 32Bwon 2 · lost 5

Claude Haiku 4.5won 1 · lost 4

DeepSeek V4 Flashwon 1 · lost 4

GPT OSS 20Bwon 1 · lost 4

MiMo V2.6 Prowon 1 · lost 4

MiMo V2.6 Flashwon 1 · lost 4

Gemini 3.8 Flashwon 1 · lost 4

Qwen3.8 Maxwon 1 · lost 4

Grok 4won 1 · lost 4

GPT-4.1won 1 · lost 5

Qwen3 235B A22B Instructwon 1 · lost 5

GPT-5.2won 1 · lost 5

GPT-5 nanowon 1 · lost 5

Claude Opus 4.7won 1 · lost 5

GPT-5.4won 1 · lost 5

GPT-5won 1 · lost 5

Claude Sonnet 4won 1 · lost 5

Claude Opus 5.5won 1 · lost 5

GPT-5.5won 1 · lost 5

Claude Sonnet 4.5won 1 · lost 6

GPT-5 miniwon 1 · lost 6

GPT-5.6 Terrawon 0 · lost 5

Grok 4.6won 0 · lost 5

O3won 0 · lost 5

Grok 4.7won 0 · lost 5

Qwen3.5 397B A17Bwon 0 · lost 5

Grok 4.5won 0 · lost 5

Claude Opus 4.8won 0 · lost 5

DeepSeek V4.1 Flashwon 0 · lost 6

§ 3 · Sources

Where the numbers come from

4 publications, 4 figures. Every one links to the page it was read from.

GDPval-AA ↗1 measureread 2026-11-11

GDPval-AA -17

BFCL ↗1 measureread 2026-11-11

BFCL v4 41.03%

Kagi LLM Benchmark ↗1 measureread 2026-11-11

Kagi LLM Benchmark 49.1%

Vectara Hallucination Leaderboard ↗1 measureread 2026-11-11

Vectara · Factual consistency 94.6%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=qwen3-14b)](https://publicai.io/model-index/m/qwen3-14b)