LLM API Pricing Index. September 2026
Updated 2026-09-30
Official prices per million tokens for 52 models from 8 providers, side by side, with APIsRouter's price where it is lower. A new edition every month, free to cite and download.
| Models | 52 |
|---|---|
| Providers | 8 |
| Lowest blended price | Gemini 2.5 Flash-Lite, $0.175 |
| Highest blended price | GPT-5.5 Pro, $67.50 |
| Median blended price | $3 |
Every model, per million tokens
Each provider's own list price in US dollars. Blended is three parts input to one part output. The APIsRouter column shows our input and output price where it is lower on every token type the provider lists. Anthropic's cache-write prices are in the files.
| Provider | Model | Input | Cached input | Output | Blended (3:1) | APIsRouter |
|---|---|---|---|---|---|---|
| OpenAI | GPT-6 Astra (gpt-6-astra) | $10 | $1 | $50 | $20 | $8 / $40 |
| OpenAI | GPT-6.1 Sol (gpt-6.1-sol) | $2 | $0.10 | $10 | $4 | $1.50 / $8 |
| OpenAI | GPT-6 Sol (gpt-6-sol) | $2 | $0.20 | $10 | $4 | $1.50 / $8 |
| OpenAI | GPT-6 Luna (gpt-6-luna) | $0.10 | $0.01 | $0.50 | $0.20 | $0.08 / $0.40 |
| OpenAI | GPT-5.6 Sol (gpt-5.6-sol) | $4 | $0.40 | $20 | $8 | $3 / $16 |
| OpenAI | GPT-5.6 Terra (gpt-5.6-terra) | $2 | $0.20 | $12 | $4.50 | $1.50 / $9.50 |
| OpenAI | GPT-5.6 Luna (gpt-5.6-luna) | $0.20 | $0.02 | $1.20 | $0.45 | $0.15 / $0.95 |
| OpenAI | GPT-5.5 (gpt-5.5) | $5 | $0.50 | $30 | $11.25 | $4 / $24 |
| OpenAI | GPT-5.5 Pro (gpt-5.5-pro) | $30 | $180 | $67.50 | ||
| OpenAI | GPT-5.4 (gpt-5.4) | $2.50 | $0.25 | $15 | $5.625 | |
| OpenAI | GPT-5.4 mini (gpt-5.4-mini) | $0.75 | $0.075 | $4.50 | $1.6875 | |
| Anthropic | Claude Fable 5.1 (claude-fable-5-1) | $10 | $0.25 | $50 | $20 | $8 / $40 |
| Anthropic | Claude Fable 5 (claude-fable-5) | $10 | $1 | $50 | $20 | $8 / $40 |
| Anthropic | Claude Opus 5.5 (claude-opus-5-5) | $4 | $0.20 | $20 | $8 | $3 / $16 |
| Anthropic | Claude Opus 5 (claude-opus-5) | $5 | $0.50 | $25 | $10 | $4 / $20 |
| Anthropic | Claude Opus 4.8 (claude-opus-4-8) | $5 | $0.50 | $25 | $10 | $4 / $20 |
| Anthropic | Claude Opus 4.7 (claude-opus-4-7) | $5 | $0.50 | $25 | $10 | $4 / $20 |
| Anthropic | Claude Opus 4.6 (claude-opus-4-6) | $5 | $0.50 | $25 | $10 | $4 / $20 |
| Anthropic | Claude Sonnet 5.5 (claude-sonnet-5-5) | $2 | $0.20 | $10 | $4 | $1.50 / $8 |
| Anthropic | Claude Sonnet 5 (claude-sonnet-5) | $2 | $0.20 | $10 | $4 | $1.50 / $8 |
| Anthropic | Claude Sonnet 4.6 (claude-sonnet-4-6) | $3 | $0.30 | $15 | $6 | $2.40 / $12 |
| Anthropic | Claude Haiku 4.5 (claude-haiku-4-5-20251001) | $1 | $0.10 | $5 | $2 | $0.80 / $4 |
| Gemini 3.8 Flash (gemini-3.8-flash). After December 31, 2026: $1.50 / $7.50 | $0.75 | $0.075 | $3.75 | $1.50 | $0.60 / $3 | |
| Gemini 3.7 Flash (gemini-3.7-flash). After December 31, 2026: $1.50 / $7.50 | $0.75 | $0.075 | $3.75 | $1.50 | $0.60 / $3 | |
| Gemini 3.6 Flash (gemini-3.6-flash). After December 31, 2026: $1.50 / $7.50 | $0.75 | $0.075 | $3.75 | $1.50 | $0.60 / $3 | |
| Gemini 3.5 Flash (gemini-3.5-flash) | $1.50 | $0.15 | $9 | $3.375 | $1.20 / $7 | |
| Gemini 3.5 Flash-Lite (gemini-3.5-flash-lite) | $0.30 | $0.03 | $2.50 | $0.85 | ||
| Gemini 3.1 Pro Preview (gemini-3.1-pro-preview). Prompts over 200,000 tokens: $4 / $18 | $2 | $0.20 | $12 | $4.50 | $1.50 / $9.50 | |
| Gemini 3.1 Flash-Lite (gemini-3.1-flash-lite) | $0.25 | $0.025 | $1.50 | $0.5625 | $0.20 / $1.20 | |
| Gemini 3 Flash Preview (gemini-3-flash-preview) | $0.50 | $0.05 | $3 | $1.125 | $0.40 / $2.40 | |
| Gemini 2.5 Pro (gemini-2.5-pro). Prompts over 200,000 tokens: $2.50 / $15 | $1.25 | $0.125 | $10 | $3.4375 | ||
| Gemini 2.5 Flash (gemini-2.5-flash) | $0.30 | $0.03 | $2.50 | $0.85 | $0.24 / $2 | |
| Gemini 2.5 Flash-Lite (gemini-2.5-flash-lite) | $0.10 | $0.01 | $0.40 | $0.175 | ||
| xAI | Grok 4.7 (grok-4.7). Prompts of 200,000 tokens or more: $4 / $12 | $2 | $0.50 | $6 | $3 | |
| xAI | Grok 4.6 (grok-4.6). Prompts of 200,000 tokens or more: $4 / $12 | $2 | $0.50 | $6 | $3 | $1.50 / $4.50 |
| xAI | Grok 4.5 (grok-4.5). Prompts of 200,000 tokens or more: $4 / $12 | $2 | $0.30 | $6 | $3 | |
| xAI | Grok 4.3 (grok-4.3). Prompts of 200,000 tokens or more: $2.50 / $5 | $1.25 | $0.20 | $2.50 | $1.5625 | |
| xAI | Grok Build 0.1 (grok-build-0.1). Prompts of 200,000 tokens or more: $2 / $4 | $1 | $0.20 | $2 | $1.25 | |
| DeepSeek | DeepSeek V4.1 Flash (deepseek-flash). Peak hours: $0.30 / $1.20 | $0.15 | $0.003 | $0.60 | $0.2625 | $0.13 / $0.50 |
| DeepSeek | DeepSeek V4 Pro (deepseek-v4-pro). Peak hours: $1.32 / $3.96 | $0.66 | $0.022 | $1.98 | $0.99 | $0.55 / $1.50 |
| Alibaba Cloud | Qwen3.8-Max (qwen3.8-max) | $2 | $6 | $3 | $1.40 / $4 | |
| Alibaba Cloud | Qwen3.8-Flash (qwen3.8-flash) | $0.15 | $0.47 | $0.23 | $0.10 / $0.30 | |
| Moonshot AI | Kimi K3 (kimi-k3) | $3 | $0.30 | $15 | $6 | $2.50 / $13.50 |
| Moonshot AI | Kimi K2.7 Code (kimi-k2.7-code) | $0.95 | $0.19 | $4 | $1.7125 | $0.85 / $3.50 |
| Moonshot AI | Kimi K2.7 Code Highspeed (kimi-k2.7-code-highspeed) | $1.90 | $0.38 | $8 | $3.425 | |
| Moonshot AI | Kimi K2.6 (kimi-k2.6) | $0.95 | $0.16 | $4 | $1.7125 | $0.85 / $3.50 |
| Z.ai | GLM-5.3 (glm-5.3) | $1.40 | $0.26 | $4.40 | $2.15 | $1.20 / $3.50 |
| Z.ai | GLM-5.3-Flash (glm-5.3-flash) | $0.15 | $0.03 | $0.50 | $0.2375 | $0.13 / $0.45 |
| Z.ai | GLM-5.3-FlashX (glm-5.3-flashx) | $0.37 | $0.075 | $1.25 | $0.59 | |
| Z.ai | GLM-5.2 (glm-5.2) | $1.40 | $0.26 | $4.40 | $2.15 | $1.20 / $3.50 |
| Z.ai | GLM-5.1 (glm-5.1) | $1.40 | $0.26 | $4.40 | $2.15 | $1.20 / $3.50 |
| Z.ai | GLM-4.7 (glm-4.7) | $0.60 | $0.11 | $2.20 | $1 |
Cite this index
Quote any figure with a link to this page. Each edition also has its own dated CSV, JSON and Markdown files with the exact numbers. APIsRouter LLM API Pricing Index, September 2026. https://apisrouter.com/llm-api-pricing Free to reuse under CC BY 4.0: credit APIsRouter and link to this page.
How the index is built
- Prices come from each provider's public pricing page, checked on the date shown for that provider: standard pay-as-you-go text prices in US dollars per million tokens. Batch and priority prices are left out, and Alibaba Cloud's are its international (Singapore) prices.
- Where a provider charges more for long prompts, at peak hours or after an introductory period, the table shows the base price, and the notes under the provider or the model give the rest.
- Blended price weighs input three to one against output, to rank models on one number. What you pay depends on your own mix of input and output.
- The APIsRouter column shows our price for the same model where it is lower than the provider's on every token type the provider lists, and stays empty otherwise.
- A new edition comes out every month, with its own dated files. This edition covers OpenAI, Anthropic, Google, xAI, DeepSeek, Alibaba Cloud, Moonshot AI and Z.ai.
One key for the 39 models with an APIsRouter price
Both ways to pay fill one balance, spent at the APIsRouter prices above.
FAQ
Which LLM API is the cheapest?
Of the 52 models in this edition, by blended price, Gemini 2.5 Flash-Lite from Google, at $0.10 / $0.40 per million input and output tokens ($0.175 blended). Blended weighs input three to one against output; the table and the files give every price so you can rank by your own mix.
Which LLM API is the most expensive?
Of the 52 models in this edition, by blended price, GPT-5.5 Pro from OpenAI, at $30 / $180 per million input and output tokens ($67.50 blended).
How is the blended price worked out?
Three parts input to one part output: three times the input price plus the output price, divided by four, per million tokens.
How often is the index updated?
Every month. Each edition has its own dated CSV, JSON and Markdown files.
Can I use this data?
Yes, under CC BY 4.0: credit APIsRouter and link to https://apisrouter.com/llm-api-pricing. The files carry the same prices as the table.
What is the APIsRouter column?
Our price for the same model, shown where it is lower than the provider's on every token type the provider lists: 39 of the 52 models in this edition. One APIsRouter key calls them all, and you can start on a $2 trial credit.