HEAD-TO-HEAD MODEL ARENA

Claude Sonnet 5.5 vs Gemini 4 Argon

Side-by-side benchmark differential across coding SWE-bench, human battle Arena ELO, throughput, and API token spend.

Claude Sonnet 5.5

Anthropic

  • 🏆 Arena Elo: 1415 pts
  • 💻 SWE-bench: 98.5/100
  • 🤖 Agentic: 98/100
  • 💵 Input: $2.00/1M
  • 💵 Output: $10.00/1M

Gemini 4 Argon

Google

  • 🏆 Arena Elo: 1395 pts
  • 💻 SWE-bench: 97.2/100
  • 🤖 Agentic: 96/100
  • 💵 Input: $1.50/1M
  • 💵 Output: $6.00/1M