Atlas

Models

← All models

Claude 3.7 Sonnet

Anthropic · 2025-02-24

Claude 3.7 Sonnet is Anthropic's first hybrid-reasoning Sonnet model, offering near-instant responses plus an optional extended-thinking mode for harder problems. It targets coding, multimodal tasks, and agentic work.

Benchmark scores

BenchmarkScore
A-Fantasia45.3
A-Fantasia Backwards Spelling35.0
A-Fantasia Chess33.0
A-Fantasia Cube Rotation68.0
AA-LCR60.7
AA-Omniscience Index0.3
Agentic Tool Use (Chat)56.3
Agentic Tool Use (Enterprise)65.3
AidanBench2170
Aider Polyglot64.9
AIME 2024 (avg@8)49.2
AIME 2024-202544.6
AIME 202549.5
AIME 2025 (avg@8)40.0
ARC-AGI-128.6
ARC-AGI-20.9
Arena Text — Business, Management & Financial Ops (No Style Control)1273
Arena Text — Chinese (No Style Control)1298
Arena Text — Creative Writing (No Style Control)1317
Arena Text — English (No Style Control)1311
Arena Text — Entertainment, Sports & Media (No Style Control)1300
Arena Text — Exclude Ties (No Style Control)1241
Arena Text — Exclude Ties (Style Controlled)1347
Arena Text — Expert (No Style Control)1300
Arena Text — French (No Style Control)1303
Arena Text — French (Style Controlled)1388
Arena Text — German (No Style Control)1280
Arena Text — Hard Prompts (No Style Control)1315
Arena Text — Hard Prompts English (No Style Control)1322
Arena Text — Instruction Following (No Style Control)1325
Loading Atlas data…