tokenstat
tokenstat

Compare

GPT-5 Mini vs Claude Haiku 4.5 Thinking

List rates and scores as of . Lower $/MTok is cheaper. Higher scores rank better. Input edge: GPT-5 Mini Output edge: GPT-5 Mini

Model A

GPT-5 Mini

OpenAI
openai/gpt-5-mini
reasoningtools
  • List input$0.04
  • List output$0.29
  • Context400K
  • Offers46
  • AA intelligence16.8
  • AA coding15.6
  • AA agentic6.8

On plans: Copilot Business, Copilot Enterprise, Copilot Free, Copilot Pro

Model B

Claude Haiku 4.5 Thinking

Anthropic
anthropic/claude-4.5-haiku
reasoningtools
  • List input$0.14
  • List output$0.71
  • Context200K
  • Offers85
  • AA intelligence16.9
  • AA coding43.9
  • AA agentic8

On plans: Claude Enterprise, Claude Max $100, Claude Max $200, Claude Pro

Score sources on /benchmarks.