The Model Index
Up to four models

o3 vs Qwen3

On the index (GPT-4 = 100), o3 scores 147 and Qwen3 133. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

o3Qwen3
Next to o3
Presets
o3
147131–163
WeightsClosed
ReleasedApr 2025
Price in / out$2.0 / $8.0
Context200K
Core tests taken6 of 9
Qwen3
133110–156
WeightsOpen
ReleasedApr 2025
Price in / out$0.12 / $0.50
Context41K
Core tests taken0 of 9

o3 and Qwen3 are too close to call (14 points apart, within the uncertainty), at 16× the price.

Test by testthird-party results

o3Qwen3
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProMATHSMATH 500MATH 500MGSMMGSMCODINGLiveCodeBenchLiveCodeBenchPROFESSIONAL WORKLegalBenchLegalBenchTaxEval v2TaxEval v2
Every result as a table
Testo3Qwen3
Reasoning & knowledge
GPQA Diamond (Vals AI)84.1%70.2%
MMLU Pro85.6%81.3%
Maths
MATH 50094.6%94.6%
MGSM91.8%92.5%
Coding
LiveCodeBench83.9%70.6%
Professional work
LegalBench83.8%80.2%
TaxEval v274.6%70.7%
Task length · METR39 sec

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG