o3

OpenAI · openai/o3

ProviderOpenAI
Familyo
Typellm-reasoning
Statusactive
Released2025-04-16
Updated2025-04-16

Reported benchmark scores

These scores carry one collection date for the whole card (2026-04) and a source list rather than a source per score, so they cannot be attributed individually. They are shown as reported and are not verified evidence. Benchmark pages carry the reviewed, dated evidence where it exists.
BenchmarkCatalogue standingScore
Aider Polyglot Benchmarkunassessed80.5
AIME 2025unassessed91.5
Arena Elo — Codingunassessed1410.0
Arena Elo — Hard Promptsunassessed1439.7
Arena Elo — Mathunassessed1430.0
Arena Elo — Overall (Text)unassessed1400.0
Arena Elo — Style Controlunassessed1431.2
Arena Elo — Visionunassessed1217.6
GPQA Diamondunassessed87.7
HumanEvalunassessed92.8
IFEvalunassessed91.0
LiveCodeBenchunassessed58.3
MATH-500unassessed96.7
MedQAunassessed96.1
MMLU-Prounassessed84.1
MultiPL-E: C#unassessed87.8
MultiPL-E: Juliaunassessed69.8
MultiPL-E: Kotlin (unconfirmed)unassessed82.1
MultiPL-E: Luaunassessed65.2
MultiPL-E: Perlunassessed62.1
MultiPL-E: PHPunassessed85.1
MultiPL-E: Runassessed67.5
MultiPL-E: Rubyunassessed77.5
MultiPL-E: Scalaunassessed71.8
MultiPL-E: Swiftunassessed75.8
SWE-bench Agentunassessed62.8
SWE-bench Verifiedunassessed71.7
τ-benchunassessed65.5
Terminal-Benchunassessed58.2

Data

This card as JSON · Edit on GitHub