Skip to content
ModelsChatgrok-4.7

grok-4.7

Google Vertex

Chat

grok-4.7

Grok 4.7 is SpaceXAI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and managing long context, and it improves on its predecessor at professional knowledge work such as drafting documents and presentations. The model was trained with a longer reinforcement learning run weighted toward problems that take many hours to complete, and natively understands the Grok Bot harness for conversational tasks. It ships with a new safeguard stack that pairs strong jailbreak resistance with low refusal rates for legitimate cybersecurity and biology work. SpaceXAI's reported benchmark results use the xhigh reasoning effort.

Input / output per 1M
Tiered
$2.00 / $6.00
0 < Token ≤ 200K
Context Window
524K
tokens
Max output
500,000
tokens
Released
2026-09-21
Chat

Provider pricing

per 1M tokens

The same model is offered by several providers. Prices below are per provider; select a row to switch provider.

ProviderBillingInput / 1MOutput / 1MCache read / 1MContextMax output

Tiered

$2.00$6.00$0.50

524K

500,000

Pricing

USD per 1M tokens

Tiered by context window; prices below are from the current provider, per 1M tokens.

Token rangeInput / 1MOutput / 1MCache read / 1M

0 < Token ≤ 200K

$2.00$6.00$0.50

200K < Token ≤ 524.3K

$4.00$12.00$1.00

Performance

Metrics unavailable

Data for this section is coming soon.

Uptime

Metrics unavailable

Data for this section is coming soon.

One API for the world’s leading AI models.

© 2026 Cloudsome. All rights reserved.