Free tool · Prices checked
AI API cost calculator: GPT vs Claude vs Gemini
Enter how big your prompts and replies are and how many calls you make, and see what each model costs per request, per day and per month at standard list prices. Every price comes from the provider's official pricing page (linked below) and was checked on October 22, 2026.
Rough guide: 1,000 tokens is about 750 English words. Month = 30 days. Standard tier only: no batch, caching, regional or fast-mode pricing.
| Model | Provider | Per request | Per day | Per month |
|---|---|---|---|---|
| Turn on JavaScript to use the calculator, or use the price table below. | ||||
Prices move every week. We tell you when.
Hotfix Daily is a free weekday email with five things worth your attention in AI, agents, vibe coding and game dev, including price changes and model retirements. Join free and get 13 ready-to-paste Starter Prompts right away.
The prices behind the calculator
Standard paid-tier text prices in US dollars per 1 million tokens, copied from each provider's official page on October 22, 2026. Retired and limited-access models are left out.
| Model | Provider | Input / 1M | Output / 1M | Notes | Source |
|---|---|---|---|---|---|
| gpt-6-astra | OpenAI | $10.00 | $50.00 | $20.00 / $75.00 when the prompt is over 272K tokens | OpenAI pricing |
| gpt-6.1-sol | OpenAI | $2.00 | $10.00 | $4.00 / $15.00 when the prompt is over 272K tokens | OpenAI pricing |
| gpt-6-luna | OpenAI | $0.10 | $0.50 | $0.20 / $0.75 when the prompt is over 272K tokens | OpenAI pricing |
| gpt-5.6-sol | OpenAI | $4.00 | $20.00 | $8.00 / $30.00 when the prompt is over 272K tokens OpenAI says its promotional pricing runs at least through Nov 21, 2026. | OpenAI pricing |
| gpt-5.6-terra | OpenAI | $2.00 | $12.00 | $4.00 / $18.00 when the prompt is over 272K tokens | OpenAI pricing |
| gpt-5.6-luna | OpenAI | $0.20 | $1.20 | $0.40 / $1.80 when the prompt is over 272K tokens | OpenAI pricing |
| gpt-5.4-mini | OpenAI | $0.75 | $4.50 | OpenAI pricing | |
| gpt-5.4-nano | OpenAI | $0.20 | $1.25 | OpenAI pricing | |
| Claude Fable 5.1 | Anthropic | $10.00 | $50.00 | Anthropic pricing | |
| Claude Opus 5.5 | Anthropic | $4.00 | $20.00 | Anthropic pricing | |
| Claude Sonnet 5.5 | Anthropic | $2.00 | $10.00 | Anthropic pricing | |
| Claude Sonnet 5 | Anthropic | $2.00 | $10.00 | Anthropic pricing | |
| Claude Haiku 5.5 | Anthropic | $0.10 | $0.50 | $0.50 / $2.50 when the prompt is over 100K tokens | Anthropic pricing |
| Claude Haiku 4.5 | Anthropic | $1.00 | $5.00 | Anthropic pricing | |
| gemini-4-argon | $2.00 | $10.00 | Price through Nov 30, 2026; Google lists $4.00 in / $20.00 out from Dec 1, 2026. | Google pricing | |
| gemini-3.8-flash | $0.75 | $3.75 | Price through Dec 31, 2026; Google lists $1.50 in / $7.50 out from Jan 1, 2027. | Google pricing | |
| gemini-3.5-flash-lite | $0.30 | $2.50 | Google pricing | ||
| gemini-3.1-flash-lite | $0.25 | $1.50 | Text/image/video input price. | Google pricing | |
| gemini-3.1-pro-preview | $2.00 | $12.00 | $4.00 / $18.00 when the prompt is over 200K tokens | Google pricing | |
| gemini-2.5-pro | $1.25 | $10.00 | $2.50 / $15.00 when the prompt is over 200K tokens | Google pricing | |
| gemini-2.5-flash | $0.30 | $2.50 | Text/image/video input price. | Google pricing |
Sources (checked October 22, 2026): OpenAI API pricing · Anthropic Claude API pricing · Gemini Developer API pricing
Ways to cut the bill
- Batch it. All three providers list Batch API prices at 50% off for work that doesn't need an instant reply.
- Cache your long prefix. A cached system prompt or document is billed at a fraction of the input price on all three (see each source page for the exact rate).
- Route by step. Send classification and simple tool calls to a small model and keep the big one for planning and hard debugging.
- Cap output. Output tokens cost about 5 times more than input on most of these models (up to about 8 times on some), and reasoning tokens count as output.
FAQ
How is an AI API call priced?
By tokens. Providers charge one rate per million input tokens (your prompt, system prompt, tool definitions and history) and a higher rate per million output tokens (the reply, including reasoning or thinking tokens). Cost per request = input tokens x input price / 1,000,000 + output tokens x output price / 1,000,000.
Why might my real bill differ from this calculator?
This uses standard list prices only. Batch processing is 50% cheaper at OpenAI, Anthropic and Google; prompt caching makes repeated input much cheaper; some models charge more above a prompt-length threshold; and OpenAI charges 10% more on regional data-residency endpoints for newer models. Tokenizers also differ: Anthropic says Claude 4.7 and later models produce about 30% more tokens for the same text.
Where do these prices come from?
From each provider's official API pricing page, copied by hand on October 22, 2026. Each row links to its source. Prices change often, so check the source before you commit to a budget.
Found this useful? Get price changes in your inbox, plus 13 free Starter Prompts. Related guides: Which AI coding agent should you use? · Which AI agent framework should you use?
How we built this: prices were copied by hand from the official pages linked above on October 22, 2026. Something out of date? Email hello@hotfixdaily.com. No provider paid to be listed.
Get the next issue in your inbox
Five things worth your attention in AI, agents, vibe coding and game dev, every weekday. Plus 13 free Starter Prompts and the AI Builder Starter Stack PDF when you join.