Breaking

xAI Releases Grok 4.7 for Coding, Agents and Knowledge Work

October 4, 2026 at 15:52 EDT

Foundation Models
AI Agents
Software Dev & Coding

xAI has released Grok 4.7, calling it its most capable model for coding and knowledge work and positioning it against other frontier systems on cost, speed and long-running task performance. The model is available through the xAI API, Grok Build, Cursor, third-party coding harnesses, model routers and cloud platforms, with standard API pricing starting at $2 per million input tokens and $6 per million output tokens.

Sep. 21, 2026 · xAI

Grok 4.7 Targets Coding and Agent Work

xAI released a larger frontier model with a 500,000-token context window, lower listed prices than several comparable systems, and benchmark gains over Grok 4.6.

Input price per 1M tokens

Output price per 1M tokens

500K

Token context window

1,657

AA Briefcase v1.1

Output price comparison

Grok 4.7

$20

GPT-5.6 Sol Max

$50

Fable 5.1 Max

Columns are proportional to listed output price per 1 million tokens.

Why developers may test it

Lower listed token prices, Cursor and Grok Build availability, stronger scores than Grok 4.6, and agent-focused training make it practical for coding trials.

What remains uncertain

Real task cost depends on prompt length, cache behavior, reasoning effort and use of the faster variant; Grok 4.7 does not lead every benchmark.

Agent workflow focus

Long task→Self-check→Tool use→Finished work

The model is trained for coding, knowledge work and the Grok Bot harness rather than only short chat sessions.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.