Free Tool
LLM API Cost Calculator: Compare Claude, GPT-6, and Gemini 4 Pricing
Enter your typical request size and monthly volume to see what the same workload would cost on each model. Everything runs in your browser; nothing you type is sent anywhere.
Not sure how many tokens your prompt is? Estimate from text
Paste a sample prompt. This uses a rough rule of about 4 characters per token for English text. Real tokenizers differ by model and language, so treat it as a ballpark.
Estimated cost by model
| Model | Input / Output per 1M tokens | Cost per request | Cost per month |
|---|
How this is calculated
Cost per request = (uncached input tokens × input price + cached input tokens × cached price + output tokens × output price) ÷ 1,000,000. Monthly cost multiplies that by your request volume. Prices are standard-tier list prices in US dollars per million tokens, checked against each vendor's own pricing page on the date above. They don't include batch discounts, fast or priority modes, regional or data-residency uplifts, taxes, or cloud-marketplace markups. Where a vendor charges more for very long prompts, the calculator applies that rate once your input passes the threshold.
Prices and sources
- Claude Opus 5.5: $4.00 input / $20.00 output / $0.20 cached. Standard tier. Anthropic's pricing page notes that Claude 4.7 and later models use a tokenizer that produces about 30% more tokens for the same text than older ones, so the same content can cost more than the per-token price suggests. Sources: Anthropic pricing docs. Our explainer.
- Claude Sonnet 5.5: $2.00 input / $10.00 output / $0.10 cached. Standard tier. Same tokenizer note as Opus 5.5: about 30% more tokens for the same text than older Claude models, per Anthropic. Sources: Anthropic pricing docs. Our explainer.
- GPT-6 Astra: $10.00 input / $50.00 output / $1.00 cached. Standard tier, short context. OpenAI's pricing page lists higher long-context rates (2x input, 1.5x output) but the threshold wasn't stated on the page. Fast mode costs about 2x. Sources: OpenAI API pricing, OpenAI launch post. Our explainer.
- GPT-6.1 Sol: $2.00 input / $10.00 output / $0.10 cached. Standard tier. OpenAI lists long-context rates of 2x input/cached and 1.5x output; the 272K-token threshold is the one OpenAI documents for GPT-6 Sol and DataCamp reports for 6.1 Sol. Sources: OpenAI API pricing, OpenAI launch post, DataCamp. Our explainer.
- GPT-6 Sol: $2.00 input / $10.00 output / $0.20 cached. Standard tier. OpenAI's model page says prompts over 272K input tokens are billed at 2x input and cached rates and 1.5x output. OpenAI now points to GPT-6.1 Sol as the newer Sol model. Sources: OpenAI GPT-6 Sol model page, SiliconANGLE. Our explainer.
- GPT-6 Luna: $0.10 input / $0.50 output / $0.01 cached. Standard tier, short context. OpenAI lists higher long-context rates but the threshold wasn't stated on the page. Sources: OpenAI API pricing, SiliconANGLE. Our explainer.
- Gemini 4 Argon (intro price): $2.00 input / $10.00 output / $0.10 cached. Introductory price through November 30, 2026, then $4 / $20 from December 1. Per Google's Gemini API pricing page, API use isn't available on the free tier or the Google AI Pro plan. Sources: Google Gemini API pricing, Google Argon rollout post (Oct 13). Our explainer.
- Gemini 4 Argon (from Dec 1): $4.00 input / $20.00 output / $0.20 cached. Price from December 1, 2026, after the introductory period ends. Per Google's Gemini API pricing page, API use isn't available on the free tier or the Google AI Pro plan. Sources: Google Gemini API pricing, Google Argon rollout post (Oct 13). Our explainer.
Heads up: AI pricing changes often, and several of these models launched only weeks ago. We update this table by hand and last checked it against the vendors' pricing pages on the date above, so confirm on the vendor's page before you commit to a budget. Also keep in mind that different models count tokens differently: Anthropic says Claude 4.7 and later models use a tokenizer that produces about 30% more tokens for the same text, so the same prompt can cost more on Claude than the per-token price alone suggests. Gemini 4 Argon's introductory price runs through November 30, 2026, and per Google's Gemini API pricing page it isn't available for API use on the free tier or the Google AI Pro plan.