xAI Frontier Available in FlowHunt

Grok 4.7

Grok 4.7 is a Frontier-tier AI model developed by xAI, which now operates under the SpaceXAI name, and released in September 2026. It is a proprietary closed-weight model, available through the Grok API, Grok Build, Cursor, and other third-party tools and platforms. The 500K-token context window is large enough to hold sizeable codebases and lengthy technical documents.

Read more
Grok 4.7 was released on September 21, 2026 and replaces Grok 4.6. It uses a new, larger base model and was trained with a longer reinforcement learning run on harder tasks, weighted toward problems that take many hours to complete. The company says it verifies its own work better, manages longer context better, and natively understands the Grok Bot harness for conversational tasks. It is served at the same price and speed as Grok 4.6.

Benchmarks

Each score appears once, with its rank among tracked models and who measured it.

Compared with Frontier peers on the same scale
  • Grok 4.7this model
  • GPT-6.1 Solthis model leads on 0 of 4
  • Claude Opus 5.5this model leads on 0 of 5
Verified independent evaluator, counted in the ranking Self-reported published by xAI, shown but not ranked
Overall

AA Intelligence Index

Independent composite of agent, coding, knowledge and science tests by Artificial Analysis (v4).

Source ↗

Grok 4.7 46.4 GPT-6.1 Sol 51.8 Claude Opus 5.5 57.6
#10of 45 Verified
Coding

DeepSWE

113 long-horizon software tasks written from scratch, graded by hidden tests.

Source ↗

Grok 4.7 71.0% GPT-6.1 Sol No score Claude Opus 5.5 74.2%
Not ranked Self-reported
Agents

Terminal-Bench 4.0

Multi-step professional tasks done by an agent in a command-line terminal.

Source ↗Also: Self-reported 37.6% (chart on launch page; column mapping inferred from the price rows (Grok 4.7, Grok 4.6, GPT-5.6 Sol, Fable 5.1))

Grok 4.7 25.8% GPT-6.1 Sol 56.1% Claude Opus 5.5 59.6%
#16of 36 Verified

Humanity's Last Exam

Expert-written questions across 100+ subjects, no tools.

Source ↗

Grok 4.7 43.1% GPT-6.1 Sol 52.9% Claude Opus 5.5 61.4%
#17of 44 Verified

ARC-AGI-2

Novel visual puzzles that cannot be memorised.

Source ↗

Grok 4.7 61.4% GPT-6.1 Sol 94.2% Claude Opus 5.5 93.3%
#17of 27 Verified

What to Use Grok 4.7 For

Long-Running Coding Tasks

It was trained on problems that take many hours to complete and is available inside Cursor and Grok Build, which makes it a fit for extended agentic coding sessions.

Knowledge Work

The company positions it for professional work such as creating documents and presentations, and for tasks done by lawyers, nurses, and financial analysts.

Agentic Tool Calling

xAI's API documentation describes it as its flagship model for code and everything else, with agentic tool calling and configurable reasoning effort.

Security Work

It ships with a new safeguard stack. Select cybersecurity partners get invite-only access to its red-team capabilities for defense research.

Strengths and limitations

Strengths

  • Long-running coding and agentic tasks
  • Professional knowledge work
  • Configurable reasoning effort
  • Same price and speed as Grok 4.6, per xAI

Limitations

  • Closed-weight, so it cannot be self-hosted or fine-tuned on private data
  • Smaller context window than the 1M-token models

Use it in FlowHunt

Use Grok 4.7 in any flow, next to your own tools and knowledge sources.

  1. 1Open a flow, or create one from a template
  2. 2Add an AI Agent or Generator component
  3. 3Pick Grok 4.7 in the model selector

Related AI Models

About xAI California, USA · Founded 2023 · x.ai

xAI is an AI company founded in 2023 by Elon Musk. It develops the Grok family of models, which are available through the Grok app, the X platform and an API.

x.ai ↗

Browse the Full AI Model Leaderboard

Frequently asked questions