model comparison

vs

Grok 4.7 vs Grok 4.6

two xAI models, compared on listed price, real workload cost, context and capabilities.

the short answer

  • Grok 4.7 and Grok 4.6 list the same input and output prices.
  • Grok 4.6 retired on 4 October 2026; its numbers are kept for reference.

side-by-side specs

specGrok 4.7Grok 4.6
providerxAIxAI
input price$2/M$2/M
output price$6/M$6/M
cached input$0.5/M$0.5/M
blended price (3:1)$3/M$3/M
context window500k tokens500k tokens
inputtext, imagestext, images
structured outputyesyes
weightsclosedclosed
releasedSeptember 21, 2026August 2026
availabilityliveretired 4 October 2026

same work, both models

workloadGrok 4.7Grok 4.6
one chat reply3,000 tokens in, 500 outsame$0.009$0.009
summarise a 100-page report50,000 tokens in, 1,500 out, one request$0.11$0.11
1,000 classification calls500 tokens in, 20 out, per call$1.12$1.12
coding-agent session2M tokens in across ~60 turns, 60,000 out, no cache$4.36$4.36
same session, cached85% of input read from the prompt cache, the rest written to it$1.81$1.81

at listed rates. token counts are typical sizes, not measurements.

Grok 4.7 pricing notes

  • once a prompt reaches 200K tokens, every rate doubles for the whole request ($4 input, $12 output, $1 cached); the listed price is the sub-200K rate.

Grok 4.6 pricing notes

  • rates double once a request reaches 200K input tokens; the listed price is the sub-200K rate.

when to pick which

Grok 4.7

xAI's model for coding and knowledge work, launched September 2026 as the successor to Grok 4.6 at the same price and speed. it takes text and images, and reasoning is always on.

pick it if

  • you want xAI's top model for coding and multi-hour terminal tasks
  • your prompts repeat: cached input bills at a quarter of the input rate
  • you need text and image input in one 500K window

skip it if

  • Grok Build 0.1 fits your work: half the input, a third the output price
  • you need a non-reasoning mode: reasoning on Grok 4.7 cannot be disabled

Grok 4.6

Grok 4.6 was xAI's August 2026 flagship, the successor to Grok 4.5, taking text and images with reasoning that could not be switched off. it was retired in favour of Grok 4.7, at the same price and context window.

pick it if

  • you wanted adjustable reasoning effort up to xhigh on a 500K window
  • you reused long prompts: cache hits billed at a quarter of the input rate

skip it if

  • you are choosing today; Grok 4.7 replaced it at the same price

faq

which is cheaper, Grok 4.7 or Grok 4.6?

Grok 4.7 and Grok 4.6 list the same input and output prices. on a coding-agent session (2M tokens in, 60,000 out) Grok 4.7 costs $4.36 and Grok 4.6 $4.36.

try either model

switch between models inside the same octomind session.

prices from the catalog of 2026-11-18