Supported models and pricing
Cost estimates in the workspace, the CLI, the API, and the extensions come from this table. Prices are list prices for direct API access, in US dollars per 1 million tokens. They are estimates: your bill also depends on caching, batch discounts, output tokens, and your contract.
- Last checked: . Last change: .
- 44 priced models. Each row shows when its price was last verified. A price not re-verified for 30 days is labeled stale.
Where the numbers come from
No AI provider offers an API for its prices, and their pricing pages are not built to be read by programs. A scheduled job therefore reads the public OpenRouter models index, which lists each provider's list price and context window, compares it with this table every day, and proposes a change when something differs. A person reviews every change before it is published. Each provider below links to its own pricing page, which is the authority: if it disagrees with this table, the provider is right.
Nothing here is estimated or filled in. A model with no confirmed price shows a dash, and the workspace asks you to enter your own rate. An implausible change (more than five times up or down in a day) is held for manual confirmation instead of applied.
OpenAI
| Model | Input / 1M | Output / 1M | Context window | Token count | Price verified |
|---|---|---|---|---|---|
GPT-6.1 Sol Progpt-6.1-sol-pro | $2 | $10 | 1,050,000 | approximate | |
GPT-6.1 Solgpt-6.1-sol | $2 | $10 | 1,050,000 | approximate | |
GPT-6 Luna Progpt-6-luna-pro | $0.1 | $0.5 | 1,050,000 | approximate | |
GPT-6 Lunagpt-6-luna | $0.1 | $0.5 | 1,050,000 | approximate | |
GPT-6 Sol Progpt-6-sol-pro | $2 | $10 | 1,050,000 | approximate | |
GPT-6 Solgpt-6-sol | $2 | $10 | 1,050,000 | approximate | |
GPT-6 Astragpt-6-astra | $10 | $50 | 1,050,000 | approximate | |
GPT-6 Astra Progpt-6-astra-pro | $10 | $50 | 1,050,000 | approximate | |
GPT-5.6 Luna Progpt-5.6-luna-pro | $0.2 | $1.2 | 1,050,000 | approximate | |
GPT-5.6 Lunagpt-5.6-luna | $0.2 | $1.2 | 1,050,000 | approximate | |
GPT-5.6 Terra Progpt-5.6-terra-pro | $2 | $12 | 1,050,000 | approximate | |
GPT-5.6 Terragpt-5.6-terra | $2 | $12 | 1,050,000 | approximate | |
GPT-5.6 Sol Progpt-5.6-sol-pro | $2 | $10 | 1,050,000 | approximate | |
GPT-5.6 Solgpt-5.6-sol | $2 | $10 | 1,050,000 | approximate | |
GPT-5.5 Progpt-5.5-pro | $30 | $180 | 1,050,000 | approximate | |
GPT-5.5gpt-5.5 | $5 | $30 | 1,050,000 | approximate | |
GPT-5.4 Nanogpt-5.4-nano | $0.2 | $1.25 | 400,000 | approximate | |
GPT-5.4 Minigpt-5.4-mini | $0.75 | $4.5 | 400,000 | approximate | |
GPT-5.4 Progpt-5.4-pro | $30 | $180 | 1,050,000 | approximate | |
GPT-5.4gpt-5.4 | $2.5 | $15 | 1,050,000 | approximate | |
GPT-4.1gpt-4.1 | $2 | $8 | 1,047,576 | exact | |
GPT-4.1 minigpt-4.1-mini | $0.4 | $1.6 | 1,047,576 | exact | |
GPT-4ogpt-4o | $2.5 | $10 | 128,000 | exact | |
GPT-4o minigpt-4o-mini | $0.15 | $0.6 | 128,000 | exact | |
GPT-4 (legacy)gpt-4 | — | — | 8,192 | exact | — (no reference price) |
Anthropic
| Model | Input / 1M | Output / 1M | Context window | Token count | Price verified |
|---|---|---|---|---|---|
Claude Haiku 5.5claude-haiku-5-5 | $0.1 | $0.5 | 1,000,000 | approximate | |
Claude Fable 5claude-fable-5 | $10 | $50 | 1,000,000 | approximate | |
Claude Opus 4.8claude-opus-4-8 | $5 | $25 | 1,000,000 | approximate | |
Claude Opus 4.7claude-opus-4-7 | $5 | $25 | 1,000,000 | approximate | |
Claude Opus 4.6claude-opus-4-6 | $5 | $25 | 1,000,000 | approximate | |
Claude Fable 5.1claude-fable-5-1 | $10 | $50 | 1,000,000 | approximate | |
Claude Opus 5.5claude-opus-5-5 | $4 | $20 | 1,000,000 | approximate | |
Claude Sonnet 5.5claude-sonnet-5-5 | $2 | $10 | 1,000,000 | approximate | |
Claude Opus 5claude-opus-5 | $5 | $25 | 1,000,000 | approximate | |
Claude Sonnet 5claude-sonnet-5 | $2 | $10 | 1,000,000 | approximate | |
Claude Sonnet 4.6claude-sonnet-4-6 | $3 | $15 | 1,000,000 | approximate | |
Claude Haiku 4.5claude-haiku-4-5 | $1 | $5 | 200,000 | approximate |
| Model | Input / 1M | Output / 1M | Context window | Token count | Price verified |
|---|---|---|---|---|---|
Gemini 3.8 Flashgemini-3.8-flash | $0.75 | $3.75 | 1,048,576 | approximate | |
Gemini 3.7 Flashgemini-3.7-flash | $0.75 | $3.75 | 1,048,576 | approximate | |
Gemini 3.6 Flashgemini-3.6-flash | $0.75 | $3.75 | 1,048,576 | approximate | |
Gemini 3.5 Flash Litegemini-3.5-flash-lite | $0.3 | $2.5 | 1,048,576 | approximate | |
Gemini 3.5 Flashgemini-3.5-flash | $1.5 | $9 | 1,048,576 | approximate | |
Gemini 3.1 Flash Litegemini-3.1-flash-lite | $0.25 | $1.5 | 1,048,576 | approximate | |
Gemini 2.5 Progemini-2.5-pro | $1.25 | $10 | 1,048,576 | approximate | |
Gemini 2.5 Flashgemini-2.5-flash | $0.3 | $2.5 | 1,048,576 | approximate |
Meta
| Model | Input / 1M | Output / 1M | Context window | Token count | Price verified |
|---|---|---|---|---|---|
Llama 3.3 70Bllama-3.3-70b | — | — | 128,000 | approximate | — (no reference price) |
Mistral
| Model | Input / 1M | Output / 1M | Context window | Token count | Price verified |
|---|---|---|---|---|---|
Mistral Largemistral-large | — | — | 128,000 | approximate | — (no reference price) |
Exact and approximate token counts
Counts are exact where the provider publishes its tokenizer (OpenAI's GPT-4 family). For other models there is no offline tokenizer, so the count is an approximation from a similar encoding, and every surface labels it that way. For an exact figure, use the provider's own token-counting endpoint.
The same data from the API
GET /api/v1/models returns this table as JSON, including priceCurrency, priceUnit,priceSource, priceSourceUrl, pricingUrl, priceVerifiedAt, andpriceEffectiveFrom for each model. See the HTTP API reference.