Claude Code

🔒 Closed

Anthropic's agentic CLI with deep shell integration, multi-file editing, extended thinking, and sub-agent spawning.

💰 Included with Claude subscription · cli

76.9Overall Score

Non-Gameable Scoring

Scores are derived from established benchmarks, adjusted for harness-specific performance across four dimensions: Coding, Reasoning, Tool Use, and Autonomy.

Each dimension starts from public benchmark data and applies harness-specific modifiers based on tool integration, context handling, and orchestration quality. The overall score is a weighted composite that penalizes narrow optimization.

Model	Overall	Coding	Reasoning	Tool Use	Autonomy
Claude Opus 4.6	76.9	75.38	89.61	64.53	89.38
Claude Sonnet 4.6	60.1	69.37	57.55	62.33	88.22
Claude Sonnet 4.5	41.2	57.6	49.76	33.21	65.46
Claude Opus 4.5	38.5	54.62	49.82	24.77	63.24
Claude Opus 4.1	29.2	42.61	46.03	14.46	43.11
Claude Opus 4	21.9	40.19	31.61	8.55	29.38