Claude Sonnet 3

Anthropic · anthropic/claude-3-sonnet-20240229

ProviderAnthropic
Familyclaude-sonnet
Typevlm
Statusactive
Released2024-03-04
Updated2024-03-04

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
AI2D (AI2 Diagrams)unassessed88.1
AlpacaEvalunassessed34.9
Arena Elo — Codingunassessed1190.0
Arena Elo — Hard Promptsunassessed1280.5
Arena Elo — Mathunassessed1180.0
Arena Elo — Overall (Text)unassessed1200.0
Arena Elo — Style Controlunassessed1279.8
Arena Elo — Visionunassessed1017.7
BBQ (Bias Benchmark for QA)unassessed85.2
ChartQAunassessed80.8
DocVQAunassessed89.3
GPQA Diamondunassessed45.5
GSM8Kunassessed92.3
HELM Safetyunassessed89.5
HumanEvalunassessed73.0
IFEvalunassessed75.5
MATH-500unassessed55.7
MathVistaunassessed50.5
MGSM (Multilingual Grade School Math)unassessed83.5
MMLU-Prounassessed60.1
MMMUunassessed53.1
ToxiGenunassessed92.8

Data

This card as JSON · Edit on GitHub