OpenAI vs Anthropic Pricing — 2026 Cost Comparison

The full GPT family vs the full Claude family, per million tokens, side-by-side.

OpenAI and Anthropic are the two leading commercial LLM providers in 2026. Both offer a three-tier ladder — Flagship, Balanced, Budget — but the pricing strategies are very different.

OpenAI ranges from GPT-5 nano at $0.05/$0.40 per million tokens up to GPT-5.5 Pro at $30/$180. Anthropic ranges from Claude Haiku 4.5 at $1/$5 up to Claude Opus 4.7 at $5/$25. OpenAI is cheaper at the budget end; Anthropic is cheaper at the flagship end. The calculator below lets you put your actual usage against both families.

Models included

GPT-5.5 (OpenAI) — $5.00 input / $30.00 output per 1M tokens · 1M context window
GPT-5 (OpenAI) — $1.25 input / $10.00 output per 1M tokens · 400K context window
GPT-5 mini (OpenAI) — $0.25 input / $2.00 output per 1M tokens · 400K context window
Claude Opus 4.7 (Anthropic) — $5.00 input / $25.00 output per 1M tokens · 1M context window
Claude Sonnet 4.6 (Anthropic) — $3.00 input / $15.00 output per 1M tokens · 1M context window
Claude Haiku 4.5 (Anthropic) — $1.00 input / $5.00 output per 1M tokens · 200K context window

Frequently asked questions

Is OpenAI or Anthropic cheaper in 2026?

It depends on tier. At the budget end OpenAI is cheaper (GPT-5 nano $0.05/$0.40 vs Haiku 4.5 $1/$5). At the flagship end Anthropic is slightly cheaper (Opus 4.7 $5/$25 vs GPT-5.5 $5/$30 and Pro $30/$180).

Which family has better budget models?

OpenAI. GPT-5 nano ($0.05/$0.40) and GPT-5 mini ($0.25/$2.00) sit well below Claude Haiku 4.5 ($1/$5) on price. For high-volume budget workloads, OpenAI wins by 5–20×.

Which family has the best frontier model?

Task-dependent. GPT-5.5 leads on coding benchmarks; Opus 4.7 leads on long-context reasoning. They tie on input price ($5/M); Opus is 17% cheaper on output.