Skip to content
← All models
xAI

Grok 4.6 benchmark and release status

xAI Unconfirmed Proprietary, availability pending Official website
PENDING
No score and no rank

Elon Musk publicly described Grok 4.6 as arriving around August 7, 2026. As checked on August 8, xAI's public model documentation still recommends Grok 4.5 for coding and does not list Grok 4.6. The target is not treated as a launch, so this entry has no Vibe Coding Index score or rank.

Pending benchmark with no placement. Grok 4.6 stays outside the numbered rankings until complete source results support a Vibe Coding Index score.

Evidence confidence

Pending evidence

0% core coverage

The model remains outside verified rankings until all core scores are available.

0/3
Core scores
0
Evidence sources
Last measured

Catalog capabilities

What the model supports

— context
Capability details will appear after the next OpenRouter catalog refresh.
Input
Not listed
Output
Not listed
Knowledge cutoff
Not listed
Cache read price
— per million tokens

Release and evidence tracker

What we know about Grok 4.6

This page separates confirmed facts from details that still need an official source or independent measurement.

Unconfirmed

Confirmed information

  • Musk described an around-August-7 target and said the planned model would use a 1.5-trillion-parameter base with improved supervised fine-tuning and reinforcement learning.
  • xAI's public model documentation, last updated July 9, recommends Grok 4.5 for coding.
  • No Grok 4.6 model card, public API identifier, pricing page, or independently verifiable benchmark profile was found in this check.

What is still pending

  • An xAI model announcement or public API model identifier
  • Confirmed pricing, context limits and availability
  • Verified Intelligence, Coding and Agentic benchmark inputs

Closest current comparison

Grok 4.5 is the newest publicly documented xAI coding model tracked in the catalog.

Source record

Checked August 8, 2026

No estimated scores

VCI breakdown

Final Intelligence, Coding and Agentic indices are pending.

Suite radar

Pending

Suite scores

Intelligence
General reasoning & knowledge (Intelligence Index)
Pending
Coding
Code generation & software tasks (Coding Index)
Pending
Agentic
Multi-step tool use & agentic workflows (Agentic Index)
Pending

Recorded score history

How Grok 4.6 has moved

All benchmark history →
The first historical point will appear after the next completed benchmark refresh.
SpaceXAI

Company pulse

Latest from SpaceXAI

Up to ten recent public posts from the company and its leadership, linked to the originals on X.

Loading recent posts…

Public posts are cached briefly for speed.

Model benchmark FAQ

Grok 4.6 for vibe coding and agentic engineering

Answers update from this model’s current sourced scores, practical rankings, and verified pricing.

Has Grok 4.6 been released?

Not confirmed. August 7 was described as an around-date by Elon Musk, but this page requires an xAI public model listing or announcement before it marks the model released.

Does Grok 4.6 have benchmark scores?

No verified profile is available. Claims or predictions are not used in the Vibe Coding Index.

Where does Grok 4.6 rank?

It does not rank yet. The model remains unranked until its release and all three independently checkable index inputs are available.