API Pricing

LLM API pricing compared — OpenAI, Gemini & Anthropic, text and image, per token and per image.

Standard pay-as-you-go rates, per 1M tokens for text models and per image for generation models. Prices in USD. Synced 08 Jun 2026. marks the current flagship for each provider. Output prices are highlighted — they typically cost 3–6× the input rate and dominate real-world spend.

Text models · per 1M tokens

ModelInputCached inOutput
GPT-5.51.05M ctx · reasoning$5.00$0.50$30.00
GPT-5.5 Promax-precision reasoning$30.00$180.00
GPT-5.4$2.50$0.25$15.00
GPT-5.4 mini$0.75$4.50
GPT-5.4 nano$0.20$1.25
GPT-4.11M ctx$2.00$0.50$8.00
GPT-4.1 mini1M ctx$0.40$0.10$1.60
GPT-4.1 nanocheapest in lineup$0.10$0.40
o3-proreasoning$20.00$80.00
GPT-4o minilegacy · 128K$0.15$0.075$0.60

Image models · per image · 1024×1024 · Low → High quality

ModelLowHighImage tokens in / out
GPT Image 2current flagship$0.006$0.211$8 / $30 per 1M
GPT Image 1.5$0.009$0.20$8 / $32 per 1M
GPT Image 1 Minicheapest$0.005$0.052
GPT Image 1deprecating 23 Oct 2026$0.011$0.25
DALL·E 2 / 3removed from API 12 May 2026discontinued

Discount levers & caveats

Nine ways prices move from the sticker rates — expand each for the detail.

Pricing changes frequently — always confirm on the official pages before committing budget: openai.com/api/pricing · ai.google.dev/pricing · docs.claude.com/en/docs/about-claude/pricing