AI Model Comparisons

Side-by-side ratings for the models people actually choose between, with pricing and context windows checked against each vendor's own documentation.

• Claude Code vs Cursor Claude Code runs inside Cursor, so this is rarely an either/or
• Claude Code vs GitHub Copilot Copilot Pro can host Claude Code, which makes the price gap misleading
• Claude Haiku 4.5 vs Claude Sonnet 5 Haiku 4.5 is a generation behind Sonnet 5, not a smaller version of it
• Claude Opus 5 vs Claude Sonnet 5 Sonnet 5 is the default; Opus 5 earns its 2.5x price on long agentic work
• Claude Opus 5 vs GPT-5.6 Sol Sol is 20% cheaper per token, but Claude's tokenizer changes the arithmetic
• Cursor vs GitHub Copilot Copilot is an extension you add; Cursor asks you to switch editors
• Gemini 3.7 Flash vs GPT-5.6 Sol Gemini 3.7 Flash is five times cheaper, but this is a Flash tier against a flagship

Looking for something broader? Browse AI tools by task.