Noometry

Model comparison

o3 vs QwQ-32B

o3 is the stronger model overall, scoring 47.5 to 39.8 on the Noometry Index.

Last verified . 27 shared benchmarks.

o3 OpenAI

47.5

Rank #61 Confirmed

QwQ-32B Alibaba (Qwen)

39.8

Rank #159 Confirmed

Summary

  • They share 27 benchmarks with published results for both. o3 scores higher in 8 categories and QwQ-32B in 0 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where o3 leads 54.6 to 37.2.
  • The biggest single-benchmark swing is Aider Polyglot: 81.3% for o3 and 20.9% for QwQ-32B.
  • QwQ-32B has downloadable open weights; the other is API-only.

Side by side

o3QwQ-32B
ProviderOpenAIAlibaba (Qwen)
Noometry Index47.539.8
Released2025-04-162024-11-28
WeightsProprietaryOpen
Context window200K—
Max output100K—
Input $ / M tokens$2—
Output $ / M tokens$8—
Results tracked6336

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding o3 leads

o3: 46.8 (#64), QwQ-32B: 35.4 (#226)

Benchmarko3QwQ-32B
Aider Polyglot81.3%20.9%
LMArena Coding14081333
SWE-bench Verified62.3%—
SWE-bench Verified (bash only)58.4%—
GSO8.8%—
WeirdML52.4%—
BigCodeBench Instruct—44.6%
LiveBench Coding—72.2%
BigCodeBench Complete—54.4%
CadEval74%—
ALE-Bench933.55—

Agentic & Tool Use Not comparable

o3: 34.5 (#44), QwQ-32B: —

Benchmarko3QwQ-32B
Berkeley Function Calling Leaderboard63%—
GDPval30.8%—
DeepResearch Bench45.2%—
OSWorld23%—
LMArena Search1144—
METR Time Horizons65.4%—

Reasoning o3 leads

o3: 32.0 (#78), QwQ-32B: 23.7 (#174)

Benchmarko3QwQ-32B
Chess Puzzles38%5%
LMArena Hard Prompts14021325
Epoch Capabilities Index146.86137.6
ForecastBench62.558.3
ARC-AGI-26.5%—
SimpleBench53.1%—
Kagi LLM Benchmark67.6%—
ARC-AGI-160.8%—
CritPt1.4%—
EnigmaEval13.1%—
LiveBench Reasoning—83.5%
Mystery Game Puzzles29%—
DTBench84.8%—
LiveBench Data Analysis—65%
LMCA39.7%—
LiveBench—72%

Math o3 leads

o3: 50.2 (#58), QwQ-32B: 38.0 (#143)

Benchmarko3QwQ-32B
OTIS Mock AIME 2024-202584.4%59.2%
LMArena Math14261359
FrontierMath (Tiers 1-3)33.3%—
Omni-MATH71.4%—
LiveBench Math—77.8%
MATH Level 597.8%—
FrontierMath (Feb 2025 set)18.7%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge o3 leads

o3: 54.6 (#52), QwQ-32B: 37.2 (#158)

Benchmarko3QwQ-32B
GPQA Diamond81.8%65.3%
Confabulations14.4%15.6%
LMArena Expert14021324
Humanity's Last Exam20.3%—
SimpleQA Verified49.4%—
MMLU-Pro85.9%—
GPQA (HELM)75.3%—

Multimodal Not comparable

o3: 41.4 (#36), QwQ-32B: —

Benchmarko3QwQ-32B
LMArena Vision1214—
GeoBench74%—
VPCT52%—

Multilingual o3 leads

o3: 51.7 (#105), QwQ-32B: 44.8 (#176)

Benchmarko3QwQ-32B
LMArena Non-English14011305
LMArena Chinese14371378
LMArena French14301336
LMArena German14201313
LMArena Japanese14031262
LMArena Korean13701279
LMArena Russian14061297
LMArena Spanish13951354

Instruction Following Too close to call

o3: 72.8 (#127), QwQ-32B: 72.6 (#137)

Benchmarko3QwQ-32B
LMArena Instruction Following13681297
LiveBench Instruction Following—81.8%
IFEval86.9%—

Long Context o3 leads

o3: 53.3 (#6), QwQ-32B: 49.0 (#11)

Benchmarko3QwQ-32B
Fiction.LiveBench88.9%83.3%
LMArena Longer Query13721308
CL-bench17.8%—

Writing & Preference o3 leads

o3: 63.5 (#64), QwQ-32B: 50.6 (#180)

Benchmarko3QwQ-32B
LMArena Text14101329
LMArena Creative Writing13591288
Short-Story Creative Writing83.9%80.2%
EQ-Bench Creative Writing16761257
LMArena Multi-Turn14051314
WildBench86.1%—
LiveBench Language—51.4%

Frequently asked questions

Is o3 better than QwQ-32B?

o3 is the stronger model overall, scoring 47.5 to 39.8 on the Noometry Index.

Is o3 or QwQ-32B better for coding?

o3 scores higher on coding benchmarks: 46.8 versus 35.4 in the Noometry coding category.

How many benchmarks do o3 and QwQ-32B share?

27 benchmarks have published results for both models. o3 has 63 scored results on Noometry and QwQ-32B has 36.

Related comparisons

Go deeper