Updated daily at 05:00 UTC

Reasoning-only

QwQ-32B

By Alibaba Group, in the QwQ family.

Released: 6 Mar 2025
Status: Current
Context: 131.1K
Max output: —
Input per 1M tokens: —
Output per 1M tokens: —
Open weights: Yes
License: apache-2.0
Parameters: 32B
Knowledge cutoff: —
Reasoning: Yes
Modalities: text → text

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Category scores
Category	Score	Rank	Coverage
Knowledge & reasoning	0.0	Not ranked	1 of 4
Coding & software engineering	20.9	Not ranked	1 of 5
Human preference	40.9	Not ranked	1 of 5

Benchmark results

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning

Knowledge & reasoning results
Benchmark	Metric	Value	Provenance	Setting	Evaluated	Source
Chess Puzzles	Puzzles solved	5.0%	Independent	—	28 Aug 2026	Epoch AI

Scientific reasoning

Scientific reasoning results
Benchmark	Metric	Value	Provenance	Setting	Evaluated	Source
GPQA Diamond	Accuracy	65.3%	Independent	—	28 Aug 2026	Epoch AI

Mathematics

Mathematics results
Benchmark	Metric	Value	Provenance	Setting	Evaluated	Source
OTIS Mock AIME 2024-2025	Accuracy	59.2%	Independent	—	28 Aug 2026	Epoch AI

Coding & software engineering

Coding & software engineering results
Benchmark	Metric	Value	Provenance	Setting	Evaluated	Source
Aider Polyglot	Percent correct	20.9%	Third party	—	—	https://aider.chat/docs/leaderboards/ via Epoch AI

Long context

Long context results
Benchmark	Metric	Value	Provenance	Setting	Evaluated	Source
Fiction.LiveBench	% correct at 16K tokens (Epoch's column)	83.3%	Third party	—	—	Fiction.live leaderboard via Epoch AI

Human preference

Human preference results
Benchmark	Metric	Value	Provenance	Setting	Evaluated	Source
Text Arena (arena.ai)	Arena score (Elo scale)	1336	Independent	—	25 Sep 2026	Arena
Text Arena (arena.ai)	Votes	25080	Independent	—	25 Sep 2026	Arena

Other QwQ versions

Sources

QwQ-32B: benchmarks, price and specs · supermarket.dev