xAI

Grok models

Version comparison

Last verified Oct 24, 2026. Figures that could not be confirmed on an official vendor page are marked unverified — we do not guess.

Full Grok comparison

Which one should I use?

xAI’s models page: for code and chat, use Grok 4.7. It is described as the most capable model they have built.
Grok 4.3 is the cheaper text model in the current pricing table ($1.25 / $2.50 per 1M below 200k prompt tokens) with a larger 1M context window.
Grok 4.6 is still listed at the same token price as 4.7 — treat it as the previous 4.x snapshot, not the one xAI tells you to pick.

Grok 4.7

Current

Context: 500K
Max output: Unverifiedunverified
API input / 1M: $2 / 1M
API output / 1M: $6 / 1M

Best for: Grok users and API callers who want the current flagship for coding and chat.

In the Grok apps, new customers sign up through The X Subscription (since 12 Oct 2026); existing SuperGrok plans continue for current subscribers. xAI doesn't say which consumer plan uses which API model.

Source: Grok 4.7

Grok 4.3

Current

Context: 1M
Max output: Unverifiedunverified
API input / 1M: $1.25 / 1M
API output / 1M: $2.5 / 1M

Best for: Longer documents on a budget, when you do not need 4.7’s flagship quality.

Consumer app access for Grok 4.3 was not listed on the xAI models page checked.

Source: Models & pricing

Grok 4.6

Prior gen

Context: 500K
Max output: Unverifiedunverified
API input / 1M: $2 / 1M
API output / 1M: $6 / 1M

Best for: Existing callers that have not moved to 4.7.

Consumer app access for Grok 4.6 was not listed on the xAI models page checked.

Source: Models & pricing

Version	Context	Max out	API in / 1M	API out / 1M	Source
Grok 4.7 Current`grok-4.7`	500K	Unverifiedunverified	$2 / 1M	$6 / 1M	Grok 4.7
Grok 4.3 Current`grok-4.3`	1M	Unverifiedunverified	$1.25 / 1M	$2.5 / 1M	Models & pricing
Grok 4.6 Prior gen`grok-4.6`	500K	Unverifiedunverified	$2 / 1M	$6 / 1M	Models & pricing

Grok 4.7

Grok users and API callers who want the current flagship for coding and chat.

xAI’s stated pick for code and everything else text-related.
500k token context window.
Configurable reasoning (low, medium, high, xhigh); high is the default.

Caveats

Max output tokens were not listed on the Grok 4.7 page checked.
No realtime knowledge unless you turn on search tools.
Prompts at or above 200k tokens cost double.

Prompts under 200k tokens. At or above 200k, input is $4 / 1M for the whole request.

Grok 4.3

Longer documents on a budget, when you do not need 4.7’s flagship quality.

Cheaper than Grok 4.7 on the published API table.
1M token context window — larger than 4.7’s 500k.

Caveats

xAI’s “which model” guide still points new work at 4.7.
Max output unverified.

Prompts under 200k tokens. At or above 200k, input is $2.50 / 1M.

Grok 4.6

Existing callers that have not moved to 4.7.

Still listed on the official pricing table at the same short-context rate as 4.7.

Caveats

xAI recommends Grok 4.7 for new code and chat work.

Prompts under 200k tokens. At or above 200k, input is $4 / 1M.

Catalog pages

Grok