The Model Index
Up to four models

o1 vs DeepSeek-V3

On the index (GPT-4 = 100), o1 scores 136 and DeepSeek-V3 119. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

o1DeepSeek-V3
Next to o1
Presets
o1
136120–152
WeightsClosed
ReleasedDec 2024
Price in / out–
Context–
Core tests taken4 of 9
DeepSeek-V3
119103–135
WeightsOpen
ReleasedDec 2024
Price in / out$0.26 / $1.03
Context164K
Core tests taken2 of 9

o1 leads DeepSeek-V3 by 17 points, ahead on 2 of 2 shared core tests.

Test by testthird-party results

o1DeepSeek-V3
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)SimpleBenchSimpleBenchMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)MATH (level 5)MATH (level 5)FrontierMath (2025 set)FrontierMath (2025 set)CODINGAider PolyglotAider Polyglot
Every result as a table
Testo1DeepSeek-V3
Reasoning & knowledge
GPQA Diamond (Epoch AI)76.8%56.5%
SimpleBench40.1%18.9%
Maths
AIME-style maths (OTIS mock)73.3%15.8%
MATH (level 5)94.7%64.8%
FrontierMath (2025 set)9.3%1.7%
Coding
Aider Polyglot61.7%48.4%
Task length · METR31 sec28 sec

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

This page as a chartDownload PNG