> Grok 4.6 is a closed AI model from xAI, released 12 Aug 2026. Index score 170 (GPT-4 = 100), rank 19 of 229 ranked models. API price $2.0 in / $6.0 out per million tokens, 500K context.

[Models](https://themodelindex.org/models/) /[xAI](https://themodelindex.org/labs/xai/)

# Grok 4.6

xAI

Released 12 Aug 2026

Closed

Flagship

Date verified

[Compare](https://themodelindex.org/compare/grok-4-6-vs-kimi-k3/) [Announcement](https://x.ai/news/grok-4-6)

170159–181

Index score

#19 of 229 models

4/ 8

Core tests taken

4 of 6 areas · plus 16 anchor tests

−17

Versus the frontier at release

best then: Claude Opus 5

Replaced by [Grok 4.7](https://themodelindex.org/models/grok-4-7/)

$2.0/ $6.0

Price per million tokens

input / output, list price

500K

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

GPT-5.3-Codex reached this score 6 months earlier.

## Test results · by area · tick = best by any model

Show 14 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

94%

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

66%

Coding

evidence

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

evidence

APEX-Agents

65%

Novel problems

evidence

ARC-AGI-2

67%

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#19

of 219 · 170

Epoch Capabilities Index

overall capability from many benchmarks

#19

of 239 · 156.6

LMArena · Creative writing

creative writing

#45

of 386 · 1445

LMArena · WebDev

building web apps and front ends

#18

of 123 · 1620

LMArena · Coding

coding questions in chat

#58

of 383 · 1507

LMArena · Text

overall chat quality, judged by people

#66

of 388 · 1453

LMArena · Maths

maths questions in chat

#73

of 373 · 1446

LMArena · Vision

understanding images

#36

of 149 · 1263

LiveBench · Overall

overall, with fresh questions to limit contamination

#15

of 58 · 78

LMArena · Documents

working with long documents

#23

of 41 · 1452

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

Replaced by Grok 4.7; it no longer leads any category among current models.

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

Grok

app · subscription

↗

xAI API

API

↗

#### API providers · 2 · per million tokens, in / out

x

xAI

$4.0 / $12

cheapest

A

Amazon Bedrock

$2.2 / $6.6

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

12 Aug 2026

Announcement

[x.ai ↗](https://x.ai/news/grok-4-6)

Release entry

Grok 4.6

Scored as

Grok 4.6

Artificial Analysis index

–

Output speed

–

Hugging Face

closed weights

## xAI releases · around this one

Grok Build 0.1

Closed

20 MAY 2026

Grok 4.5

Closed

8 JUL 2026

index 163

Grok 4.6

Closed

12 AUG 2026

index 170

Grok 4.7

Closed

21 SEP 2026

~178 est.

---
Source: https://themodelindex.org/models/grok-4-6/ · The Model Index · data as of 2026-10-01