Atlas
/
Models
137 sources · 855 models × 1278 benchmarks · 74,806 scores
☾
☀
Capabilities
Models
Compare
Benchmarks
Coverage
Methodology
← All models
O3 (2025-04-16)
OpenAI · 2025-04-16
Immutable o3-2025-04-16 API snapshot evaluated by ReplicationBench.
Benchmark scores
Benchmark
Score
Arena Text — Business, Management & Financial Ops (No Style Control)
1392
Arena Text — Chinese (No Style Control)
1437
Arena Text — Creative Writing (No Style Control)
1360
Arena Text — English (No Style Control)
1415
Arena Text — Entertainment, Sports & Media (No Style Control)
1374
Arena Text — Exclude Ties (No Style Control)
1393
Arena Text — Exclude Ties (Style Controlled)
1424
Arena Text — Expert (No Style Control)
1400
Arena Text — French (No Style Control)
1445
Arena Text — French (Style Controlled)
1475
Arena Text — German (No Style Control)
1419
Arena Text — Hard Prompts (No Style Control)
1402
Arena Text — Hard Prompts English (No Style Control)
1407
Arena Text — Instruction Following (No Style Control)
1368
Arena Text — Japanese (No Style Control)
1406
Arena Text — Korean (No Style Control)
1370
Arena Text — Legal & Government (No Style Control)
1429
Arena Text — Life, Physical & Social Science (No Style Control)
1428
Arena Text — Longer Query (No Style Control)
1371
Arena Text — Medicine & Healthcare (No Style Control)
1448
Arena Text — Multi-turn (No Style Control)
1405
Arena Text — Non-English (No Style Control)
1404
Arena Text — Non-English (Style Controlled)
1424
Arena Text — Occupational Mathematical (No Style Control)
1422
Arena Text — Polish (No Style Control)
1440
Arena Text — Polish (Style Controlled)
1459
Arena Text — Russian (No Style Control)
1408
Arena Text — Software & IT Services (No Style Control)
1414
Arena Text — Spanish (No Style Control)
1384
Arena Text — Writing, Literature & Language (No Style Control)
1372
Loading Atlas data…