The Model Index
Up to four models

o3 vs DeepSeek-V3 (Mar 2025)

On the index (GPT-4 = 100), o3 scores 147 and DeepSeek-V3 (Mar 2025) 124. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to o3
Presets
o3
147131–163
WeightsClosed
ReleasedApr 2025
Price in / out$2.0 / $8.0
Context200K
Core tests taken6 of 9
DeepSeek-V3 (Mar 2025)
124108–140
WeightsOpen
ReleasedMar 2025
Price in / out$0.29 / $1.14
Context164K
Core tests taken2 of 9

o3 leads DeepSeek-V3 (Mar 2025) by 23 points, ahead on 2 of 2 shared core tests, at 7.0× the price.

Test by testthird-party results

o3DeepSeek-V3 (Mar 2025)
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)Fiction.LiveBenchFiction.LiveBenchSimpleBenchSimpleBenchMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)MATH (level 5)MATH (level 5)CODINGAider PolyglotAider PolyglotWeirdMLWeirdMLAGENTSDTBenchDTBenchLMCALMCA
Every result as a table
Testo3DeepSeek-V3 (Mar 2025)
Reasoning & knowledge
GPQA Diamond (Epoch AI)81.8%67.6%
Fiction.LiveBench88.9%50.0%
SimpleBench53.1%27.2%
Maths
AIME-style maths (OTIS mock)84.4%37.8%
MATH (level 5)97.8%75.5%
Coding
Aider Polyglot81.3%55.1%
WeirdML52.4%36.1%
Agents
DTBench84.8%64.8%
LMCA39.7%15.5%
Task length · METR39 sec30 sec

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

This page as a chartDownload PNG