xAI language model
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.
01
price
input tokens
output tokens
cached input
per 1M tokens. octomind bills the listed provider price, with no markup, on every plan, including the free one (pay as you go from credits).
- 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.
02
what real work costs
| workload | cost |
|---|---|
| one chat reply 3,000 tokens in, 500 out | $0.009 |
| summarise a 100-page report 50,000 tokens in, 1,500 out, one request | $0.11 |
| 1,000 classification calls 500 tokens in, 20 out, per call | $1.12 |
| coding-agent session 2M tokens in across ~60 turns, 60,000 out, no cache | $4.36 |
| same session, cached 85% of input read from the prompt cache, the rest written to it | $1.81 |
at the listed rate. token counts are typical sizes, not measurements.
03
specs
- provider
- xAI
- context window
- 500k tokens — ≈ 750 pages of prose or 50,000 lines of code
- input
- text, images
- output
- text
- structured output
- yes — JSON schema
- reasoning mode
- yes
- prompt caching
- yes — cache hits at $0.5/M
- weights
- closed
- released
- September 21, 2026
- status
- live
- model id
grok-4.7
integration notes
- reasoning cannot be disabled; reasoning_effort accepts low, medium, high (default) or xhigh.
- presencePenalty, frequencyPenalty and stop are not supported; requests that include them return an error.
04
benchmarks
as published by the maker; not measured by octomind.
05
when to pick it
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
06
where it sits on price
blended price $3/M at 3 input tokens per output token: 31 live language models cost less, 12 cost more.
- GLM-5.3 $2.15/M
- Qwen4 Max $2.25/M
- Grok 4.7 $3/M
- Qwen3.8 Max $3/M
- Claude Sonnet 5.5 $4/M
07
use Grok 4.7 in octomind
one octohub login reaches every model in the catalog. pick this one per run, per session, or as your default.
octomind login
octomind run -m octohub:grok-4.7/model octohub:grok-4.7[model]
name = "octohub:grok-4.7"08
compare Grok 4.7
- Grok 4.7 vs Claude Sonnet 5
- Grok 4.7 vs Grok 4.6
- Grok 4.7 vs Qwen3.8 Max
- Grok 4.7 vs Claude Sonnet 5.5
- Grok Build 0.1 vs Grok 4.7
- Qwen4 Max vs Grok 4.7
09
similar models
nearest live models by listed input price.
- Claude Sonnet 5.5 $2/M input
- Gemini 3.1 Pro $2/M input
- Gemini 4 Argon $2/M input
- GPT-6.1 Sol $2/M input
10
earlier versions
retired models retained for historical reference.
11
Grok 4.7 faq
how much does Grok 4.7 cost?
Grok 4.7 lists at $2/M input, $6/M output, $0.5/M cached input per 1M tokens. a typical coding-agent session (2M tokens in, 60,000 out) comes to $4.36, or $1.81 when 85% of the input is read from the prompt cache. octomind bills the listed provider price, with no markup, on every plan, including the free one (pay as you go from credits).
what is the context window of Grok 4.7?
500k tokens, ≈ 750 pages of prose or 50,000 lines of code.
does Grok 4.7 accept images?
yes. it takes text and images as input and returns text.
is Grok 4.7 open source?
no. Grok 4.7 is a closed-weights model, available only through xAI's api and hosted gateways such as octohub.
when was Grok 4.7 released?
September 21, 2026, by xAI.
is Grok 4.7 free?
no. Grok 4.7 bills per token at $2/M input, $6/M output, $0.5/M cached input per 1M. octomind's free plan reaches it pay as you go from credits, so you pay only for the tokens you use, at the provider price. credits carry a 3% top-up fee.
how do i use Grok 4.7 in octomind?
run octomind login once, then octomind run -m octohub:grok-4.7. inside a running session, /model octohub:grok-4.7 switches to it without losing context.
use Grok 4.7 in octomind
choose a model when you start, then switch mid-session when the work changes.