Atlas

Models

← All models

Opus 4

Anthropic · 2025-05-22

Opus 4 is Anthropic's Claude 4 flagship, a hybrid-reasoning model built for complex, long-running agentic tasks, research, and coding. It supports extended thinking with tool use and was positioned as a top-tier coding model at launch.

Benchmark scores

BenchmarkScore
A-Fantasia40.3
A-Fantasia Backwards Spelling22.0
A-Fantasia Chess29.0
A-Fantasia Cube Rotation70.0
AA-LCR36.0
Aider Polyglot72.0
AIME 2024 (avg@8)48.8
AIME 2024-202541.3
AIME 2025 (avg@8)33.8
AlgoTune1.3
ARC-AGI-127.0
ARC-AGI-28.6
Arena Text — Business, Management & Financial Ops (No Style Control)1336
Arena Text — Chinese (No Style Control)1387
Arena Text — Creative Writing (No Style Control)1375
Arena Text — English (No Style Control)1377
Arena Text — Entertainment, Sports & Media (No Style Control)1349
Arena Text — Exclude Ties (No Style Control)1330
Arena Text — Exclude Ties (Style Controlled)1400
Arena Text — Expert (No Style Control)1372
Arena Text — French (No Style Control)1363
Arena Text — French (Style Controlled)1422
Arena Text — German (No Style Control)1372
Arena Text — Hard Prompts (No Style Control)1376
Arena Text — Hard Prompts English (No Style Control)1385
Arena Text — Instruction Following (No Style Control)1373
Arena Text — Japanese (No Style Control)1330
Arena Text — Korean (No Style Control)1304
Arena Text — Legal & Government (No Style Control)1372
Arena Text — Life, Physical & Social Science (No Style Control)1363
Loading Atlas data…