QwQ 32B — Benchmarks

Benchmark scores for QwQ 32B aggregated from public leaderboards (6 boards), with how it ranks among open models and how it compares with its nearest neighbours. See hardware requirements for what you need to run it.

llmrun Composite Score →

96.2

90% range 93.3–99.2 · from 5 boards

CodingNot enough data yet
AgentsNot enough data yet
MathNot enough data yet
ScienceNot enough data yet
ReasoningNot enough data yet

QwQ 32B reasoning benchmarks

Multi-step logic and problem-solving — can the model think through a hard problem rather than recall an answer.

GPQA Diamond
GPQA Diamond is a set of extremely hard, graduate-level science questions (physics, chemistry, biology) written by domain experts and filtered so that skilled non-experts with web access still fail. It measures genuine reasoning rather than memorization.
65.3%
65.3%
median 54.0%open leader 93.1%
Overall best: Gemini 4 Argon (high) 96.4%
62nd percentile · #38 of 98 open · #193 of 300 all
#38 of 98
#193 of 300
Independent
Epoch AI · Mar 2025
Fiction.LiveBench
Fiction.LiveBench is a long-context comprehension test: models answer questions that require tracking characters, events and relationships across a long story. llmrun shows the 16k-token score, the column Epoch AI reports as its headline.
83.3%
83.3%
median 61.1%open leader 86.1%
Overall best: GPT 5 (Aug 07, 2025, medium) 97.2%
94th percentile · #3 of 25 open · #11 of 58 all
#3 of 25
#11 of 58
Independent
Epoch AI · Mar 2025

QwQ 32B coding benchmarks

Writing, editing, and fixing real code — measured by whether the resulting program actually runs and passes tests.

Aider Polyglot
The Aider Polyglot benchmark measures real-world coding across several programming languages: the model edits code to solve Exercism exercises, and is scored on whether the final solution actually runs and passes the tests.
20.9%
20.9%
median 45.1%open leader 74.2%
Overall best: GPT 5 (Aug 07, 2025, high) 88.0%
35th percentile · #12 of 18 open · #57 of 69 all
#12 of 18
#57 of 69
Independent
Epoch AI · Mar 2025

QwQ 32B math benchmarks

Symbolic and quantitative problem-solving, from competition math to multi-step calculation.

AIME 2024/2025
AIME (American Invitational Mathematics Examination) is a prestigious high-school competition of hard, integer-answer problems. It's a widely-cited yardstick for multi-step mathematical reasoning, here on the 2024–2025 papers.
59.2%
59.2%
median 52.8%open leader 98.6%
Overall best: Claude Fable 5 (high) 100.0%
56th percentile · #38 of 86 open · #168 of 283 all
#38 of 86
#168 of 283
Independent
Epoch AI · Mar 2025

QwQ 32B knowledge benchmarks

Breadth and depth of factual knowledge across many subjects, paired with the reasoning to apply it.

MMLU-Pro
MMLU-Pro is a harder, cleaned-up successor to MMLU with ten answer choices and more reasoning-heavy questions across 14 subjects, measuring broad knowledge and reasoning together.
69.1%
69.1%
median 52.1%open leader 88.0%
Overall best: Gemini-3.1-Pro 91.2%
73rd percentile · #40 of 145 open · #104 of 257 all
#40 of 145
#104 of 257
Independent
TIGER-Lab

QwQ 32B instruction following benchmarks

How precisely a model obeys explicit, verifiable instructions — format, length, and content constraints — and, more broadly, how much people prefer its responses head-to-head in blind human voting.

LMArena Text
Arena (arena.ai, formerly LMArena / Chatbot Arena) ranks models by millions of blind, head-to-head human votes on real prompts. The Arena Score shown here is a Bradley-Terry rating computed with Arena's default style control, so higher means people preferred that model's answers more often — it's a relative rating, not a percentage, so compare it only to other models on this board.
1335.6
1335.6
median 1321.5open leader 1487.1
Overall best: Gemini 4 Argon (High) 1523.9
55th percentile · #85 of 187 open · #232 of 405 all
#85 of 187
#232 of 405
Arena
Arena · Oct 2026

Score vs VRAM in the 24 GB tier

QwQ 32B
Other models
Best Score for its VRAM
x: VRAM (GB) · y: llmrun Score

#16 of 35 that fit 24 GB

Each dot is a model at the best quant that fits a 24 GB GPU. The line joins the models with the best Score for their VRAM; a dot below it is beaten by a model that needs no more memory.

Full 24 GB ranking →

QwQ 32B vs its neighbours

Same boards for the nearest open models per llmrun Score and the best model that fits 24 GB. The highest score in each row is marked.

Board	QwQ 32Bthis model	Qwen3.8 27Bbest in classQ4_K_M · 18 GB	DeepSeek R1 Distill Qwen 32Bnearest by ScoreQ4_K_S · 23 GB	Qwen3 30B A3B Instruct 2507nearest by ScoreQ4_K_M · 19 GB	GPT OSS 20Bnearest by ScoreQ4_K_M · 13 GB
llmrun Score	96.2	122.1	97.2	96.7	96.3
GPQA Diamond	65.3%	—	64.1%	55.6%	60.8%
AIME 2024/2025	59.2%	—	55.6%	62.2%	65.3%
MMLU-Pro	69.1%	—	—	—	73.6%
LMArena Text	1335.6	1437.8	—	1382.3	1317.4
Compare head to head		Compare	Compare	Compare	Compare

Scores aggregated from public benchmark sources (Epoch AI: 4 boards, Arena: 1 board, TIGER-Lab: 1 board) — each linked from the benchmark pages. llmrun does not run these benchmarks. Percentiles and ranks are among open models in the llmrun catalogue; the median and leader marks refer to the same field.