Back to all models
C

Claude 3 Opus

AnthropicReleased 2024-03-04
Official Page

Anthropic's most powerful model at launch, with expert-level performance across math, coding, and reasoning.

Overall Score
78%
Cost Efficiency
100
Context Window
200K

Strengths

  • ✓Strong reasoning
  • ✓Excellent for complex tasks
  • ✓Long context window
  • ✓Vision capabilities

Weaknesses

  • ✗More expensive than 3.5 models
  • ✗Outperformed by newer models in some areas
Architecture
Transformer
Parameters
~81B activated (MoE)
Input
$15/M
Output
$75/M

Benchmark Performance

BenchmarkDescriptionScoreCategoryBar
MMLUMassive Multitask Language Understanding86.8%knowledge
HumanEvalCode generation accuracy84.9%coding
MATHMathematical reasoning87.3%math
GSM8KGrade school math87.3%math
GPQAGraduate-level science Q&A58.5%reasoning
HellaSwagCommon sense reasoning86.7%reasoning
SWE-benchSoftware engineering problems37.7%coding
ARC-CAI2 Reasoning Challenge94%reasoning
AI
ModelBench

The open benchmark platform for comparing the latest AI models and their capabilities.

Platform

  • Compare Models
  • All Models
  • Methodology

AI Labs

  • OpenAI
  • Anthropic
  • Google DeepMind
  • Meta AI

Resources

  • Documentation
  • Benchmark Guide
  • API Reference
Benchmarks are sourced from official model cards and independent evaluations. Data is for informational purposes only.
© 2026 ModelBench. Built for the AI community.