xAI: Grok 4.7
Grok 4.7 is a reasoning model from xAI released on Sep 16, 2026. It reads text, images and files and writes text, with a 500K token context window and up to 500K tokens per reply.
Show moreShow less
It costs $2 per million input tokens and $6 per million output tokens. Cached input bills at $0.50, 75% less than fresh input.
It supports tool calling, structured JSON output and prompt caching.
22 providers serve it, including Amazon Bedrock, Vercel AI Gateway and Vertex. The lowest input price is $1.60 per million tokens at Prime Inference.
On the Artificial Analysis Intelligence Index it scores 46.5, which places it in the top ten, at number 7, among the 43 models I track with independent scores.
- Modalities
- In / out price
- $2 / $6 per 1M
- Context
- 500K
- Max output
- 500K
- Released
- Sep 16, 2026
- Knowledge cutoff
- May 2026
- Tool calling
- Structured output
- Reasoning
- Prompt caching
- Open weights
- Free to use
Grok 4.7 pricing
USD, as listed on Oct 15, 2026. Most bills are input heavy, so the cached input rate matters as much as the headline price.
| Input | $2 / 1M tokens | |
|---|---|---|
| Output | $6 / 1M tokens | |
| Cached input (read) | $0.50 / 1M tokens | 75% off input |
| Prompts over 200K | $4 in, $12 out | long context rate |
| Prompts over 200K | $4 in, $12 out | long context rate |
Grok 4.7 providers
22 providers serve it, including Amazon Bedrock, Vercel AI Gateway and Vertex. The lowest input price is $1.60 per million tokens at Prime Inference. Each host sets its own price and limits, so the cheapest one is not always the one with the longest context.
| Provider | Input /M | Output /M | Cache read /M | Context | Max output |
|---|---|---|---|---|---|
| xAI First party | $2 | $6 | $0.50 | 500K | 500K |
| Amazon Bedrock | $2 | $6 | $0.50 | 500K | 500K |
| Vercel AI Gateway | $2 | $6 | $0.50 | 500K | 500K |
| Vertex | $2 | $6 | $0.50 | 524K | 500K |
| Azure | n/a | n/a | – | 500K | 500K |
| Prime Inference | $1.60 | $4.80 | – | 500K | 450K |
| 302.AI | $2 | $6 | $0.50 | 500K | 500K |
| Cloudflare AI Gateway | $2 | $6 | $0.50 | 500K | 500K |
| CrossModel | $2 | $6 | $0.50 | 500K | 500K |
| DevPass (LLM Gateway) | $2 | $6 | $0.50 | 524K | 500K |
| Eden AI | $2 | $6 | $0.50 | 500K | 500K |
| Kilo Gateway | $2 | $6 | $0.50 | 500K | 450K |
Showing 12 of 22 providers: the lab first, then the major clouds, then the cheapest resellers.
What Grok 4.7 costs per month
Pick a workload close to yours or type your own volumes.
Independent benchmarks
Measured by third parties, not reported by xAI. Each links to its source and shows the date I checked it.
A composite of 10 independent evaluations (agents 30%, coding 20%, scientific reasoning 20%, general 30%), run by Artificial Analysis on each model, version 4.3.2.
Artificial Analysis, checked Oct 10, 2026
Human preference: people compare two anonymous answers and pick the better one. Ratings from the Text Arena overall board, style control on.
Arena (formerly LMArena), checked Oct 10, 2026
Agentic coding in a real terminal: can the model, driving an agent harness, finish multi step software and ops tasks. Resolution rate.
Terminal-Bench, checked Oct 10, 2026
Abstract reasoning on visual puzzles the model has not seen, semi private set. Hard to memorise, so it tracks reasoning more than knowledge.
ARC Prize, checked Oct 10, 2026
Expert level questions across many fields, no tools allowed. The Diamond set is 1,000 questions. Few models are on it yet.
Scale Labs, checked Oct 10, 2026
Median output tokens per second measured by Artificial Analysis on the provider API, for the same model setting as its index score.
Artificial Analysis, checked Oct 10, 2026
Benchmarks xAI reports (9)
- cursorbench-4.0 (CursorBench 4.0, Grok 4.7 xHigh)46.3%
- deepswe-v1.1 (DeepSWE v1.1, high effort (marked with asterisk))71%
- eebench (EEBench, electrical engineering, xHigh)64%
- aa-briefcase-v1.1 (AA Briefcase v1.1, multi-hour office work, xHigh)1657
- terminal-bench-4.0 (Terminal-Bench 4.0, multi-hour terminal work, xHigh)37.6%
- harvey-legal-agent (Harvey Legal Agent Benchmark, xHigh)19.6%
- healthbench-professional (HealthBench Professional, clinical reasoning, xHigh)56.7%
- gdpval (GDPval, Elo, Grok 4.7 (xhigh))1695
- latchbio-biosafety (LatchBio biosafety benchmark)62.4%
Labs pick their own settings and harnesses, so compare these only within one lab.
Supported request parameters
What you can set when you call Grok 4.7 through an OpenAI compatible API.
- Tools
tools - Structured outputs
structured_outputs - Reasoning
reasoning - Temperature
temperature
Alternatives to Grok 4.7
Models that take the same inputs and call tools, closest in price, cheaper ones first.
- 500K context
Reads text, images and files. Tool calling, structured output, reasoning. Served by 29 providers.
by x-aiAug 10, 2026$2/M input tokens$6/M output tokens29 providers
- Index 45.4
Reads text, images, files and video. Tool calling, structured output, reasoning. Served by 26 providers.
by qwenJul 19, 20261M context$2/M input tokens$6/M output tokens26 providers
- 500K context
Reads text, images and files. Tool calling, structured output, reasoning. Served by 24 providers.
by x-aiJul 7, 2026$2/M input tokens$6/M output tokens24 providers
- Index 14.2
Reads text, images and files. Tool calling, structured output, reasoning. Served by 18 providers.
by mistralaiApr 29, 2026262K context$1.50/M input tokens$7.50/M output tokens18 providers
- 992K context
Reads text, images, files and video. Tool calling, structured output, reasoning. Served by 5 providers.
by qwenSep 2, 2026$2/M input tokens$6/M output tokens5 providers
- Index 48.1
Reads text, images, files, audio and video. Tool calling, structured output, reasoning. Served by 13 providers.
by metaSep 2, 20261.05M context$1.25/M input tokens$4.25/M output tokens13 providers
More from xAI
- 1K context
Generates video from text, images and files. Served by 2 providers.
by x-aiOct 1, 20262 providers
- 500K context
Reads text, images and files. Tool calling, structured output, reasoning. Served by 29 providers.
by x-aiAug 10, 2026$2/M input tokens$6/M output tokens29 providers
- 66K context
Generates images from text and images. Served by 3 providers.
by x-aiAug 7, 2026$0/M input tokens$0/M output tokens3 providers
Grok Voice STT 1.0
Transcription15K contextTurns speech into text.
by x-aiAug 4, 2026
Grok Voice TTS 1.0
Speech15K contextTurns text into spoken audio.
by x-aiJul 31, 2026
- 500K context
Reads text, images and files. Tool calling, structured output, reasoning. Served by 24 providers.
by x-aiJul 7, 2026$2/M input tokens$6/M output tokens24 providers
Grok 4.7 questions
How much does the Grok 4.7 API cost?
$2 per million input tokens and $6 per million output tokens, with cached input at $0.50. Prompts over 200K tokens bill at $4 input and $12 output. The estimator on this page turns that into a monthly bill for your own workload.
What is the context window of Grok 4.7?
500,000 tokens of input, and it can write up to 500,000 tokens in one reply. As a rough guide, a million tokens is around 750,000 English words.
Which providers serve Grok 4.7?
22 providers serve it, including Amazon Bedrock, Vercel AI Gateway and Vertex. The lowest input price is $1.60 per million tokens at Prime Inference. Prices differ by host, so the Providers table on this page lists each one with its input and output price.
Does Grok 4.7 support tool calling?
Yes. It accepts tool definitions and can return JSON that matches a schema, which is what an agent needs to call your APIs.
Is Grok 4.7 open source?
No. xAI serves it through an API only, so you cannot download the weights or host it yourself.
When was Grok 4.7 released?
Sep 16, 2026, with training data up to May 2026.
How good is Grok 4.7 compared with other models?
It scores 46.5 on the Artificial Analysis Intelligence Index and 1443 on the Arena text leaderboard. Benchmarks are a shortlist, not a verdict: test the top two or three candidates on your own tasks before you commit.
Catalog data from models.dev (MIT licensed), refreshed daily, last on Oct 15, 2026. Independent scores from my hand checked rankings, sources linked above. The summary is written from these facts.
Building on Grok 4.7?
I test your real tasks on it and the two closest alternatives, then ship the agent on whichever wins on cost and accuracy.
Book Meeting