Skip to content
The Internet Compass

AI Tools

AI Model Cost & Context Comparison

A single table for the question every team asks before choosing a model: what does it cost per million tokens, and how much context does it actually give you.

FreeNo sign-upRuns in your browser
ModelProviderInput $/1MOutput $/1MContext window
Claude Opus 5Anthropic$15.00$75.00500k
Claude Sonnet 5Anthropic$3.00$15.00500k
Claude Haiku 4.5Anthropic$1.00$5.00200k
GPT-5.1OpenAI$5.00$15.00400k
GPT-5.1 miniOpenAI$0.60$2.40400k
Gemini 3 ProGoogle$3.50$14.001,000k
Gemini 3 FlashGoogle$0.40$1.601,000k
Llama 4 MaverickMeta (hosted)$0.60$1.801,000k
Mistral Large 3Mistral$2.00$6.00256k
DeepSeek V4DeepSeek$0.30$1.20160k
10 of 10 rows

How it works

  1. 1Browse or filter by model or provider
  2. 2Compare input and output pricing side by side
  3. 3Cross-reference with the cost calculator for your actual volume

Who this is for

  • Shortlisting models before an evaluation
  • Sanity-checking a vendor's pricing claim
  • Explaining model cost trade-offs to a non-technical stakeholder

Frequently asked questions

Why do input and output prices differ so much?
Output generation is autoregressive and compute-heavier per token than reading input, which most providers price at 3–5x the input rate.