SpaceXAI
Grok 4.6 (high)
Grok 4.6 builds on Grok 4.5 with a particular focus on long-running agents and more ambitious interactive and visual work. It stays with complex tasks across many steps, whether researching a topic, analyzing information, working across a codebase, or turning an idea into a polished application or work artifact.
xAI model releasesModel Snapshot
TypeLanguage
ReleaseAug 12, 2026
Context500,000
Tagsreasoning, tool-use, implicit-caching, vision
Overall
Rank #35 across 687
44.3
Coding
Rank #13 across 330
76.8
Blended Price
Rank #301 across 432
$3.00/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Output Speed70.3 tok/s
First Token35.27s
Max Output450,000
Pricing
Blended$3.00/M
Input$2.00/M
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 44.3 | #35 |
| Artificial Analysis Coding Index | coding | 76.8 | #13 |
| LiveBench Mathematics | math | 92.6% | #18 |
| GPQA | reasoning | 94.9% | #5 |