AI API pricing by brand
This site tracks 55 AI API models across 6 providers. The lowest input rate is GPT-6 Luna at $0.1 per 1M tokens and the highest is GPT-6 Cyber at $20 per 1M. Each brand page lists every model that provider sells on its API, with the dated official source for each rate.
All 6 providers
| Provider | Models | Lowest input /1M | Highest input /1M | Cheapest model | Calculator |
|---|---|---|---|---|---|
| OpenAI | 21 | $0.1 | $20 | GPT-6 Luna | calculator |
| Anthropic | 13 | $0.1 | $10 | Claude Haiku 5.5 | calculator |
| 10 | $0.1 | $2 | Gemini 2.5 Flash-Lite | calculator | |
| xAI | 3 | $1 | $2 | Grok Build 0.1 | calculator |
| DeepSeek | 2 | $0.3 | $1.32 | DeepSeek Flash (V4.1) | calculator |
| Mistral | 6 | $0.1 | $0.7 | Ministral 3 3B | calculator |
What each brand page lists
Each page lists every model the provider sells on its API, cheapest first, with the input, cached-input and output rate per 1M tokens, the cost of 1,000 calls at a fixed workload, and the date the price was checked against the provider own pricing page. Nothing is copied from resellers or round-up posts, and a model whose official page shows no readable rate is left out rather than estimated.
Why browse by brand instead of one long table
The full price table answers what a token costs anywhere; a brand page answers what this provider charges for its lineup and which model to price first. The two views link to each other, and every model page keeps a single primary keyword rather than repeating it in a brand list.
How to use these pages
Pick the provider you already use, find the model closest to your token shape, then open its model page for the full tiers and worked examples. If you are still choosing, the cheapest-by-call ranking and the price-tier index cut across brands and are a faster way to shortlist.
Related
- Every model price per 1M tokens
- Models grouped by price tier
- Cheapest LLM APIs, ranked by cost per call
- Model vs model cost comparisons
- Blended cost by input:output mix
- LLM benchmark comparison
- Best LLM for coding in 2026
- How prices are verified
Questions
How many AI API providers does this site track?
This site tracks 55 models across 6 providers: OpenAI, Anthropic, Google, xAI, DeepSeek and Mistral. Each provider has a brand page listing its full lineup with the dated official source for every rate.
Are these list prices or reseller prices?
Every rate on a brand page is the list price published on the provider own pricing page, checked on the date shown in the row. Cloud platforms and resellers publish their own price lists and are not mixed into these tables.
Which provider has the cheapest model?
GPT-6 Luna from OpenAI has the lowest input rate at $0.1 per 1M tokens; the highest input rate here is GPT-6 Cyber at $20 per 1M. The brand pages show where each model sits inside its own lineup.
How often are these prices checked?
Every model row carries its own check date, and the site-wide check date is 2026-11-18. When a provider changes a price the date changes with it; the price-change log lists announced increases and promos.
Next step → See every model price per 1M tokens
Price data last checked 2026-11-18 against official sources · how we verify
Price-change alerts
One short email when a provider changes a price or a promo ends. No spam, unsubscribe any time.