model rates
Grok 4.7 API price and model rates
Grok 4.7 costs $2 per million input tokens, $6 per million output tokens,
and $0.50 per million cache-read tokens at Cursor’s published short-context
rate. PromptConduit uses that rate for grok-4.7 and
cursor-grok-4.7. The rate card on this page was updated on
.
A Grok 4.7 mode is a Cursor billing variant of the same model. Standard, Fast, 500k, and 500k Fast are four modes with their own slugs. Fast and 500k cost the same number of dollars and stay on separate rows, because a fast short prompt and a long-context prompt are different sessions. PromptConduit plan prices, the free CLI and the paid Pro and Team plans, are on the pricing page.
How much does each Grok 4.7 mode cost?
Grok 4.7 standard mode costs $2 per million input tokens, $6 per million
output tokens, and $0.50 per million cache-read tokens. Grok 4.7 Fast,
slug grok-4.7-fast, costs $4, $12, and $1, which is twice
the standard rate. Grok 4.7 500k, slug grok-4.7-500k, also
costs $4, $12, and $1, and applies when the prompt is above 256,000
tokens, up to 500,000. Grok 4.7 500k Fast, slug
grok-4.7-500k-fast, costs $6, $18, and $1.50, three times
the standard rate. A transcript that records grok-4.7-500k
is priced on the 500k row, and a transcript that records
grok-4.7-fast is priced on the Fast row. Effort names such
as high or medium reuse the price
of the matching speed and context mode. Cursor and SpaceXAI jointly
trained Grok 4.7. On Cursor’s Start plan the model runs in non-fast mode
at a fixed medium effort. PromptConduit recorded these four rows on
September 25, 2026 from Cursor’s models and pricing page.
| Model | Slug | Input | Output | Cache read |
|---|---|---|---|---|
| Grok 4.7 | cursor-grok-4.7 grok-4.7 | $2 | $6 | $0.50 |
| Grok 4.7 Fast | cursor-grok-4.7-fast grok-4.7-fast, cursor-grok-4.7-high-fast | $4 | $12 | $1 |
| Grok 4.7 500k | cursor-grok-4.7-500k grok-4.7-500k | $4 | $12 | $1 |
| Grok 4.7 500k Fast | cursor-grok-4.7-500k-fast grok-4.7-500k-fast, cursor-grok-4.7-500k-high-fast | $6 | $18 | $1.50 |
What rates does PromptConduit use for other models?
PromptConduit prices every supported model id from one snapshot dated September 25, 2026. The dollars are the public API list price per million tokens: input, output, and cache read. The CLI and the editor extension ship this same table. Rows below are the ids in that snapshot, grouped by provider.
Cursor Grok and Composer
Cursor Grok 4.7, Grok 4.6, Grok 4.5, and Composer are first-party Cursor models. On Teams and Enterprise plans they are exempt from the Cursor Token Rate of $0.25 per million tokens. A short grok-4.7 id in a Cursor transcript uses the matching cursor-grok-4.7 row.
| Model | Slug | Input | Output | Cache read |
|---|---|---|---|---|
| Grok 4.7 | cursor-grok-4.7 grok-4.7 | $2 | $6 | $0.50 |
| Grok 4.7 Fast | cursor-grok-4.7-fast grok-4.7-fast, cursor-grok-4.7-high-fast | $4 | $12 | $1 |
| Grok 4.7 500k | cursor-grok-4.7-500k grok-4.7-500k | $4 | $12 | $1 |
| Grok 4.7 500k Fast | cursor-grok-4.7-500k-fast grok-4.7-500k-fast, cursor-grok-4.7-500k-high-fast | $6 | $18 | $1.50 |
| Grok 4.6 | cursor-grok-4.6 grok-4.6 | $2 | $6 | $0.50 |
| Grok 4.6 Fast | cursor-grok-4.6-fast grok-4.6-fast, cursor-grok-4.6-high-fast | $4 | $12 | $1 |
| Grok 4.5 | cursor-grok-4.5 grok-4.5 | $2 | $6 | $0.50 |
| Grok 4.5 Fast | cursor-grok-4.5-fast grok-4.5-fast, cursor-grok-4.5-high-fast | $4 | $18 | $1 |
| Composer 2.5 | composer-2.5 | $0.50 | $2.50 | $0.20 |
| Composer 2.5 Fast | composer-2.5-fast | $3 | $15 | $0.50 |
| Composer 1 | cursor-composer-1 | $1.25 | $10 | $0.125 |
Claude
Claude rates follow Anthropic’s published API card. Cache writes default to 1.25× input for a 5-minute cache and 2× input for a 1-hour cache. Claude Fable 5.1 and Claude Mythos 5.1 cache reads are $0.25 per million tokens. Claude Opus 5.5 cache reads are $0.20 per million tokens. Cursor lists a Fast id, claude-opus-5-5-fast, on the standard Opus 5.5 price, so PromptConduit prices that id at $4 input and $20 output.
| Model | Slug | Input | Output | Cache read |
|---|---|---|---|---|
| Claude Fable 5 | claude-fable-5 | $10 | $50 | $1 |
| Claude Fable 5.1 | claude-fable-5-1 | $10 | $50 | $0.25 |
| Claude Mythos 5 | claude-mythos-5 | $10 | $50 | $1 |
| Claude Mythos 5.1 | claude-mythos-5-1 | $10 | $50 | $0.25 |
| Claude Sonnet 5 | claude-sonnet-5 | $2 | $10 | $0.20 |
| Claude Opus 5.5 | claude-opus-5-5 | $4 | $20 | $0.20 |
| Claude Opus 5 | claude-opus-5 | $5 | $25 | $0.50 |
| Claude Opus 5 Fast | claude-opus-5-fast | $10 | $50 | $1 |
| Claude Opus 4.8 Fast | claude-opus-4-8-fast | $10 | $50 | $1 |
| Claude Opus 4.8 | claude-opus-4-8 | $5 | $25 | $0.50 |
| Claude Opus 4.7 | claude-opus-4-7 | $5 | $25 | $0.50 |
| Claude Opus 4.6 | claude-opus-4-6 | $5 | $25 | $0.50 |
| Claude Opus 4.5 | claude-opus-4-5 | $5 | $25 | $0.50 |
| Claude Opus 4.1 | claude-opus-4-1 | $15 | $75 | $1.50 |
| Claude Opus 4 | claude-opus-4-0 | $15 | $75 | $1.50 |
| Claude Sonnet 4.6 | claude-sonnet-4-6 | $3 | $15 | $0.30 |
| Claude Sonnet 4.5 | claude-sonnet-4-5 | $3 | $15 | $0.30 |
| Claude Sonnet 4 | claude-sonnet-4-0 | $3 | $15 | $0.30 |
| Claude Haiku 4.5 | claude-haiku-4-5 | $1 | $5 | $0.10 |
| Claude 3.5 Haiku | claude-3-5-haiku | $0.80 | $4 | $0.08 |
Gemini
PromptConduit prices gemini-3.6-flash, gemini-3.7-flash, and gemini-3.8-flash at Google’s standard paid rates through December 31, 2026: $0.75 input, $3.75 output, and $0.075 cache read per million tokens. Google doubles those three rates on January 1, 2027. On September 25, 2026, Cursor’s model page listed Gemini 3.6 Flash at $1.50 input, $7.50 output, and $0.15 cache read, and listed Gemini 3.7 Flash and Gemini 3.8 Flash output at $3.50. A gemini-* id in PromptConduit follows Google.
| Model | Slug | Input | Output | Cache read |
|---|---|---|---|---|
| Gemini 3.8 Flash | gemini-3.8-flash | $0.75 | $3.75 | $0.075 |
| Gemini 3.7 Flash | gemini-3.7-flash | $0.75 | $3.75 | $0.075 |
| Gemini 3.6 Flash | gemini-3.6-flash | $0.75 | $3.75 | $0.075 |
| Gemini 3.5 Flash | gemini-3.5-flash | $1.50 | $9 | $0.15 |
| Gemini 3.1 Pro | gemini-3.1-pro | $2 | $12 | $0.20 |
| Gemini 3 Pro | gemini-3-pro | $2 | $12 | $0.20 |
| Gemini 3 Flash | gemini-3-flash | $0.50 | $3 | $0.05 |
| Gemini 2.5 Flash | gemini-2.5-flash | $0.30 | $2.50 | $0.03 |
OpenAI
OpenAI rows are the public API rates in the PromptConduit snapshot, including GPT-6 Sol, GPT-6 Luna, and GPT-5.3 Codex. GPT-6 Sol cache writes are $2.50 per million tokens. GPT-6 Luna cache writes are $0.125 per million tokens.
| Model | Slug | Input | Output | Cache read |
|---|---|---|---|---|
| GPT-6 Sol | gpt-6-sol | $2 | $10 | $0.20 |
| GPT-6 Luna | gpt-6-luna | $0.10 | $0.50 | $0.01 |
| GPT-5.6 Sol | gpt-5.6-sol | $4 | $20 | $0.40 |
| GPT-5.6 Terra | gpt-5.6-terra | $2 | $12 | $0.20 |
| GPT-5.6 Luna | gpt-5.6-luna | $0.20 | $1.20 | $0.02 |
| GPT-5.5 | gpt-5.5 | $5 | $30 | $0.50 |
| GPT-5.4 | gpt-5.4 | $2.50 | $15 | $0.25 |
| GPT-5.3 Codex | gpt-5.3-codex | $1.75 | $14 | $0.175 |
Grok Build, Kimi, GLM, and Muse
Grok Build 0.1 is the model behind the Grok coding CLI. It costs $1 per million input tokens, $2 per million output tokens, and $0.20 per million cache-read tokens. Grok Build 0.1 and Cursor Grok 4.7 are different models. Kimi, GLM, and Muse Spark use the provider rates stored in the same snapshot.
| Model | Slug | Input | Output | Cache read |
|---|---|---|---|---|
| Grok Build 0.1 | grok-build-0.1 | $1 | $2 | $0.20 |
| Kimi K3 | kimi-k3 | $3 | $15 | $0.30 |
| Kimi K2.7 Code | kimi-k2.7-code | $0.95 | $4 | $0.19 |
| GLM 5.2 | glm-5.2 | $1.40 | $4.40 | $0.26 |
| Muse Spark 1.3 | muse-spark-1.3 | $1.25 | $4.25 | $0.15 |
Sources checked September 25, 2026: Cursor models and pricing , Anthropic Claude API pricing , Google Gemini API pricing , OpenAI API pricing , xAI API pricing . The same snapshot is available as model-rates.json.
How does PromptConduit turn these rates into a session cost?
PromptConduit multiplies the token counts in a local session by the public API list price for that model id. Input tokens use the input rate, output tokens use the output rate, and cache-read tokens use the cache-read rate. The result is an API-equivalent dollar figure on your machine. A Cursor, Claude, Gemini, or ChatGPT subscription can include usage, so the invoice for that month is a separate number from the figure in the editor. PromptConduit plan prices are on the pricing page. Live session cost in the editor is covered in the editor extension docs.
Questions about Grok 4.7 and model rates
- How much does Grok 4.7 cost?
- Grok 4.7 costs $2 per million input tokens, $6 per million output tokens, and $0.50 per million cache-read tokens at Cursor’s published short-context rate. PromptConduit uses that rate for grok-4.7 and cursor-grok-4.7. The rate card was updated on September 25, 2026. Grok 4.7 Fast costs $4, $12, and $1. Grok 4.7 500k costs $4, $12, and $1. Grok 4.7 500k Fast costs $6, $18, and $1.50.
- What is the difference between Grok 4.7 Fast and Grok 4.7 500k?
- Grok 4.7 Fast and Grok 4.7 500k share one price: $4 per million input tokens, $12 per million output tokens, and $1 per million cache-read tokens. Fast is the speed mode on the standard context window. 500k is the long-context mode for a prompt above 256,000 tokens, up to 500,000 tokens. PromptConduit stores them as grok-4.7-fast and grok-4.7-500k so each session keeps the mode that actually ran.
- Does Grok 4.7 high have its own price?
- Grok 4.7 high uses the price of the speed and context mode it belongs to. cursor-grok-4.7-high-fast is priced as Grok 4.7 Fast at $4 input, $12 output, and $1 cache read per million tokens. cursor-grok-4.7-500k-high-fast is priced as Grok 4.7 500k Fast at $6 input, $18 output, and $1.50 cache read. On Cursor’s Start plan, Grok 4.7 runs in non-fast mode at a fixed medium effort.
- Are model rates the same as my Cursor or Claude invoice?
- PromptConduit prices a local session at the public API list price for the model id. A Cursor, Claude, Gemini, or ChatGPT subscription can include usage, so the invoice total is a separate number. On Cursor Teams and Enterprise, third-party models also add a Cursor Token Rate of $0.25 per million tokens. Grok and Composer are exempt from that fee.
- Which Gemini 3.8 Flash price does PromptConduit use?
- PromptConduit prices Gemini 3.8 Flash, Gemini 3.7 Flash, and Gemini 3.6 Flash at Google’s standard paid rates through December 31, 2026: $0.75 per million input tokens, $3.75 per million output tokens, and $0.075 per million cache-read tokens. Google doubles those rates on January 1, 2027. Cursor’s page listed Gemini 3.6 Flash at $1.50 / $7.50 / $0.15 and listed Gemini 3.7 and 3.8 output at $3.50. A gemini-* id follows Google.
- When were these model rates last updated?
- PromptConduit updated this rate card on September 25, 2026 from the published API prices on Cursor, Anthropic, Google, OpenAI, and xAI. The same snapshot ships in the PromptConduit CLI and the editor extension.