‹ PublicAI Index
The LLM benchmark aggregator.
Qwen3 32B
Alibaba · 33B · open weights
Strongest in Tool use (#23 of 81), weakest in Human preference (#179 of 354). Above par in 7 of 11 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 DeepSeek V3.2.
§ 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#90/2001/4
General intelligence51.1−17.9#79/1801/3
Safety53.2−8.2#92/3271/3
Factual grounding60.3−10.4#24/1041/1
Coding49.2−16.9#93/1791/6
Agentic coding49.1−17.6#91/1731/5
Agents47.6−18.7#168/2982/6
Tool use57.9−16.1#23/811/1
Knowledge work36.9−33.8#169/2081/2
Human preference51.6−17.2#179/3541/1
Human preference51.6−17.2#179/3541/1
§ 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 32B, left for the other.
§ 3 · Sources
Where the numbers come from
6 publications, 6 figures. Every one links to the page it was read from.
LMArena Text 1347
GDPval-AA 41
Aider polyglot 40%
BFCL v4 48.71%
Kagi LLM Benchmark 54.9%
Vectara · Factual consistency 94.1%
Badge
[](https://publicai.io/model-index/m/qwen3-32b)