LLM API pricing table
27 price rows from 5 providers, in US dollars per 1 million tokens. Prices checked . We re-check the official pricing pages every month and after each new model release.
| GPT-6 Astra | OpenAI | $10 | $50 | $1 | — | 1,050,000 | 128,000 |
|---|---|---|---|---|---|---|---|
| GPT-6.1 Sol | OpenAI | $2 | $10 | $0.1 | — | 1,050,000 | 128,000 |
| GPT-6 Luna | OpenAI | $0.1 | $0.5 | $0.01 | — | 1,050,000 | 128,000 |
| GPT-5.6 Sol | OpenAI | $4 | $20 | $0.4 | — | — | — |
| GPT-5.5 | OpenAI | $5 | $30 | $0.5 | — | — | — |
| GPT-5.4 mini | OpenAI | $0.75 | $4.5 | $0.075 | — | — | — |
| GPT-5 mini | OpenAI | $0.25 | $2 | $0.025 | — | — | — |
| GPT-4.1 | OpenAI | $2 | $8 | $0.5 | — | 1,047,576 | 32,768 |
| GPT-4o mini | OpenAI | $0.15 | $0.6 | $0.075 | — | — | — |
| Claude Fable 5.1 | Anthropic | $10 | $50 | $0.25 | — | 1,000,000 | 128,000 |
| Claude Opus 5.5 | Anthropic | $4 | $20 | $0.2 | — | 1,000,000 | 128,000 |
| Claude Sonnet 5.5 | Anthropic | $2 | $10 | $0.1 | — | 1,000,000 | 128,000 |
| Claude Haiku 5.5 (prompt up to 100K) | Anthropic | $0.1 | $0.5 | $0.01 | — | 1,000,000 | 128,000 |
| Claude Haiku 4.5 | Anthropic | $1 | $5 | $0.1 | — | — | — |
| Gemini 3.1 Pro Preview (prompt up to 200K) | $2 | $12 | $0.2 | — | 1,048,576 | 65,536 | |
| Gemini 3 Flash Preview | $0.5 | $3 | $0.05 | — | — | — | |
| Gemini 2.5 Pro (prompt up to 200K) | $1.25 | $10 | $0.125 | — | 1,048,576 | 65,536 | |
| Gemini 2.5 Flash | $0.3 | $2.5 | $0.03 | — | 1,048,576 | 65,536 | |
| Gemini 3.1 Flash-Lite | $0.25 | $1.5 | $0.025 | — | — | — | |
| Gemini 2.5 Flash-Lite | $0.1 | $0.4 | $0.01 | — | — | — | |
| Mistral Large 3 | Mistral | $0.5 | $1.5 | — | — | — | — |
| Mistral Medium 3.5 | Mistral | $1.5 | $7.5 | — | — | — | — |
| Mistral Small 4 | Mistral | $0.15 | $0.6 | — | — | — | — |
| DeepSeek V4.1 Flash (off-peak) | DeepSeek | $0.15 | $0.6 | $0.003 | — | — | — |
| DeepSeek V4.1 Flash (peak hours) | DeepSeek | $0.3 | $1.2 | $0.006 | — | — | — |
| DeepSeek V4 Pro (off-peak) | DeepSeek | $0.66 | $1.98 | $0.022 | — | — | — |
| DeepSeek V4 Pro (peak hours) | DeepSeek | $1.32 | $3.96 | $0.044 | — | — | — |
Showing 27 of 27 rows. A dash means the value is not confirmed or the provider does not publish one.
Per 1,000 requests: price per 1M tokens × tokens per request × 1,000 ÷ 1,000,000. Cached input assumes every input token hits the cache. Total is input plus output at the standard price.
Recent changes
No price changes recorded yet. The history starts with the snapshot of ; changes are listed here as soon as a later check finds a different price.
Download
Both files carry the same rows as the table and are rebuilt with the site. License: CC BY 4.0, so you may copy and reuse them with attribution.
How to cite
Link to this page and keep the check date, because prices change.
Nanocalcs. (2026). LLM API pricing table [Data set]. https://nanocalcs.com/en/llm-api-pricing/ (prices checked 2026-10-15; CC BY 4.0)<a href="https://nanocalcs.com/en/llm-api-pricing/">LLM API pricing table</a> by Nanocalcs (CC BY 4.0), prices checked 2026-10-15Source: Nanocalcs, LLM API pricing table, https://nanocalcs.com/en/llm-api-pricing/, prices checked 2026-10-15. CC BY 4.0.Method and sources
- Prices are standard-tier list prices for text input and output, in US dollars per 1 million tokens, copied from the official pricing pages below. A value we could not confirm is left empty instead of guessed.
- Cached input is the price of an input token read from the prompt cache. Cache write surcharges, cache storage fees, batch discounts, long-prompt tiers other than the condition shown, tool fees and taxes are not included.
- Context window and maximum output come from each provider's model documentation where we checked it.
- This is a reference list, not a quote. Check the official page before you commit to a budget.
Related calculators
- LLM API Cost CalculatorClaude, OpenAI (GPT) and Gemini API pricing in one calculator: enter tokens per request and requests per day to get daily and monthly API cost.
- Prompt Caching CalculatorSee how much prompt caching saves on repeated LLM calls. Cache write and read prices for Claude (5 minute and 1 hour), OpenAI and Gemini, with the break-even call count.
- Token CounterOpenAI and ChatGPT token counter (exact o200k_base, cl100k_base) with Claude and Gemini estimates. Paste text to count tokens; nothing leaves your browser.