Beta
Clasificaci贸n/Claude 3.7 Sonnet (thinking)
Anthropic

Claude 3.7 Sonnet (thinking)

por AnthropicPublicado el 2025-02-24

42.1
puntuaci贸n promedio
$3.00/1M
Precio de entrada
$15.00/1M
Precio de salida
200K tokens (~100 books)
Ventana de contexto
multimodal
Tipo

Tested on 20 benchmarks with 42.1% average. Top scores: MATH level 5 (91.2%), Fiction.LiveBench (83.3%), Lech Mazur Writing (81.1%).

Puntuaciones de benchmark

BenchmarkCategor铆aPuntuaci贸nBar
MATH level 5math91.2
Fiction.LiveBenchknowledge83.3
Lech Mazur Writingknowledge81.1
GPQA diamondknowledge73.0
GeoBenchknowledge68.0
Aider polyglotcoding64.9
OTIS Mock AIME 2024-2025math57.7
CadEvalcoding54.0
SWE-Bench Verified (Bash Only)coding52.8
DeepResearch Benchknowledge43.6
OSWorldagentic35.8
SimpleBenchreasoning35.7
The Agent Companyagentic30.9
ARC-AGIreasoning28.6
Cybenchcoding20.0
VPCTknowledge8.5
FrontierMath-2025-02-28-Privatemath4.1
GSO-Benchcoding3.8
HLEknowledge3.4
ARC-AGI-2reasoning0.9

Modelos similares