HEAD-TO-HEAD MODEL ARENA

DeepSeek R1 vs OpenAI o1

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

DeepSeek R1

DeepSeek

  • 🏆 Arena Elo: 1358 pts
  • 💻 SWE-bench: 95/100
  • 🤖 Agentic: 91/100
  • 💵 Input: $0.55/1M
  • 💵 Output: $2.19/1M

OpenAI o1

OpenAI

  • 🏆 Arena Elo: 1365 pts
  • 💻 SWE-bench: 96/100
  • 🤖 Agentic: 94/100
  • 💵 Input: $15.00/1M
  • 💵 Output: $60.00/1M