xAI · Product data

Grok price, plans and limits (November 2026)

Operational

Read 3 Nov 2026 20:00 UTC

Everything this site holds about Grok, on one page: whether it is working, what it costs, what its API charges, and what has changed. Every figure below was read from xAI's own pages, and every block says when. Where xAI publishes nothing, this page says so instead of estimating.

Download the data behind this Grok pagestatus JSON status CSV plans JSON plans CSV pricing JSON pricing CSV

Is Grok working right now?

Operationalread 3 Nov 2026 20:00 UTC

xAI describes it as: No open incidents in the vendor's incident feed.

No incidents open at the time in the stamp. Full status detail and incident list

Ninety days, one cell a day

We do not keep a day by day history yet. The filled cell on the right is the reading in the snapshot this page was built from, taken 3 Nov 2026 20:00 UTC. The empty cells mean no reading was recorded for that day, which is not the same as a day with nothing wrong. This strip fills in as the recorder runs.

Source: status.x.aiRead 3 Nov 2026 20:00 UTCSingle sourceHistory

What Grok costs

xAI's published list prices for Grok are below, with each limit in the vendor's own wording. A cell reading "Not published" means xAI prints no figure for it, which is a fact worth as much as a number.

The words in this table

Token: The unit AI companies meter and bill in. A token is a chunk of text, roughly three quarters of a word in ordinary English, so a thousand word document is about thirteen hundred tokens. Each vendor counts with its own tokenizer, so the same sentence can come to a slightly different number of tokens at one vendor than at another. Full guide
Context window: How much text a model can hold in view at one time, counted in tokens. It covers the whole exchange: what you send, any files you attach, and the answer being written. It is not the length of one message, and a vendor may allow less of it inside some features than the headline figure suggests.
Rate limit: A cap on how much you may use in a window of time: messages per few hours on a consumer plan, or requests and tokens per minute on an API. Reaching one is not an outage. The service is working and is asking you to wait.

Grok plans, prices and published limits
Plan	Monthly	Yearly per month	Context	Models
Freeindividual	Not published	Not published	Not published	Not published
Limits published for Free Free tier limits on Chat and VoicexAI FAQ: paid users who reach their weekly limit "will still have access to Grok's free tier limits on Chat and Voice" xAI's docs name a Free plan but publish no price (presumably $0, not stated) and no numeric limits. x.ai/pricing, x.ai/grok, grok.com and help.x.com all returned HTTP 403 or a JavaScript-only shell to our fetcher on 2026-09-05, so no consumer price could be read.
SuperGrokindividual	Not published	Not published	Not published	Not published
Limits published for SuperGrok Weekly usage a single usage allowance you can spend however you like"Paid SuperGrok plans raise your limits and unlock more across every product" Extra Usage Credits priced at standard ratesTop-up credits expire one year after purchase (unless otherwise stated) xAI FAQ: "Rolling out in June 2026, Grok will now use a simpler, more flexible system for paid users" with "One flexible weekly usage pool that works across all Grok products". No numeric allowance, price or model list is published on the pages we could read.
SuperGrok Heavyindividual	Not published	Not published	Not published	Not published
Limits published for SuperGrok Heavy Weekly usage allowance included with your subscriptionHigher plans give "significantly more weekly usage at a lower cost per action" than buying credits Named in xAI's Grok FAQ and Grok Business guide ("SuperGrok Heavy for upgraded licenses"); grok.com's own page markup gates some features behind "SuperGrok Heavy subscription required" and links https://grok.com/#subscribe. No official page we could fetch prints its price.

Vendor billing wording

SuperGrok: Price not readable: x.ai/pricing returned HTTP 403 to our fetcher
SuperGrok Heavy: Yearly billing exists (FAQ mentions a "SuperGrok Heavy yearly subscription"); price not readable, see notes

Grok plans:Source: docs.x.ai/grok/overviewVerified 2 Nov 2026 04:00 UTCSecond sourceHistory

Rows in this group were read from 2 pages: docs.x.ai/grok/overview, docs.x.ai/grok/faq

What it means for you

xAI publishes no paid individual price for Grok on the page we read.

Every plan here carries limits xAI publishes, which is rarer than it should be and makes this table unusually easy to plan around.

What the xAI API costs

List prices per one million tokens, exactly as xAI prints them. This is what you pay to send text to the model yourself, which is a different product from the subscription above and is billed by the word rather than by the month.

The words in this price table

Token: The unit AI companies meter and bill in. A token is a chunk of text, roughly three quarters of a word in ordinary English, so a thousand word document is about thirteen hundred tokens. Each vendor counts with its own tokenizer, so the same sentence can come to a slightly different number of tokens at one vendor than at another. Full guide
Input and output tokens: Input is the text you send, output is the text the model returns. Vendors price them separately and output is always the dearer half, which is why a long answer costs more than a long question.
Cached input: A lower price some vendors charge when you send the same opening text again soon after sending it. The vendor has already processed that part and charges less to read it back. Where a vendor publishes no cached rate, our tables say Not published rather than assuming the full price.

xAI: API list prices per million tokens in USD
Model	Input	Output	Cached input	Context	Max output
Grok Build 0.1grok-build-0.1	$1.00	$2.00	$0.20	256K	Not published
Grok 4.3grok-4.3	$1.25	$2.50	$0.20	1M	Not published
Grok 4.6grok-4.6	$2.00	$6.00	$0.50	500K	Not published
Grok 4.5grok-4.5	$2.00	$6.00	$0.30	500K	Not published

Notes on these rows

Grok Build 0.1: Aliases grok-code-fast-1, grok-code-fast, grok-code-fast-1-0825. Prompts under 200k tokens; 200k or more: $2.00 / $0.40 cached / $4.00. Context printed as "256k" on the table and "256,000 tokens" on the model page. No max output published.
Grok 4.3: Alias grok-4.3-latest. Prompts under 200k tokens; 200k or more: $2.50 / $0.40 cached / $5.00. Model page mentions a 20% Batch API discount. Context printed as "1M" on the table and "1,000,000 tokens" on the model page. No max output published.
Grok 4.6: Prompts under 200k tokens. Prompts of 200k or more: input $4.00, cached $1.00, output $12.00; "Requests whose prompt reaches the listed token threshold are billed at the higher rate for all tokens in the request." Context printed as "500k" on the table and "500,000 tokens" on the model page. No max output published.
Grok 4.5: Aliases grok-4.5-latest, grok-build-latest. Prompts under 200k tokens; 200k or more: $4.00 / $0.60 cached / $12.00. No max output published.

xAI prices:Source: docs.x.ai/docs/modelsVerified 2 Nov 2026 04:00 UTCSecond sourceHistory

US dollars per 1M input tokens

Input prices only, because every vendor publishes one for every model, so the chart has no holes in it. Output costs more at most models; the table above carries both.

What it means for you

Inside one vendor the spread is the story: Grok Build 0.1 costs $1.00 per million input tokens and Grok 4.5 costs $2.00. Choosing the model matters more than choosing the vendor.

A million tokens is roughly three quarters of a million words of English, so a single question costs a fraction of a cent. Bills grow when the same context is sent again and again, which is what an agent or a long conversation does. That is why the cached input column is the one to read before committing to a model here.

What changed at Grok

Every change we have logged at Grok, newest first, each with xAI's own page as the receipt.

26 Sep 2026Price change

Models and productsxAI

xAI Changes Grok Pricing Page at grok.com

xAI's public site at grok.com now shows a pricing page that differs from the version this site last recorded, though the vendor has not detailed the change.

21 Sep 2026Research

ResearchxAI

Paper Benchmarks Grok's Voice Agent Across Five Languages

A new paper testing voice agents across five languages found Grok leads on task completion but trails rivals on the quality of its generated speech.

17 Sep 2026Research

ResearchxAI

arXiv Paper Finds Grok Amplifies Aggression in Synthetic Cyberbullying Data

A new arXiv paper compared cyberbullying dialogues generated by Grok, GPT and LLaMA against real conversations, finding xAI's model stands out for amplifying aggressive behavior.

Everything logged at xAI · the whole stream

AI Journey is independent and is not affiliated with xAI. Every figure here is read from pages xAI publishes, and the page and time are printed beside it.