SpaceXAI
Released September 21, 2026

Grok 4.7

Grok 4.7 is SpaceXAI's flagship for coding and knowledge work, with agentic tool calling, configurable reasoning, 500k context, and minimal hallucinations.

Inputs
Text
Image
File
Outputs
Text

Model Overview

Capabilities, design details, and architectural traits

Grok 4.7 - Coding and knowledge work

Grok 4.7 is positioned for code, chat, and general knowledge work. It was trained with a longer reinforcement learning run on a harder mix of tasks, weighted toward problems that take many hours to complete, and it is built to keep working on difficult tasks rather than stopping early.

What sets Grok 4.7 apart

TraitDetail
Recommended for code and everything elsexAI recommends Grok 4.7 for code, chat, and all non-media use cases.
Configurable reasoningReasoning effort can be tuned per request, including a high-effort mode.
Encrypted reasoningOn the Responses API, Grok 4.7 always returns reasoning.encrypted_content, even when not requested.
Trained on multi-hour tasksA longer RL run on a harder task mix, weighted toward problems that take many hours.
Self-verificationDocumented improvement in checking its own work and managing longer context.
Native Grok Bot understandingTrained to natively understand the Grok Bot harness for conversational tasks and knowledge work.
500k token contextLarge context window for long-running coding and office work.
No built-in realtime knowledgeHas no knowledge of current events beyond training data unless server-side Web Search or X Search tools are enabled.

Safety and cybersecurity focus

Grok 4.7 ships with an entirely new safeguard stack. Select cybersecurity partners get invite-only access to its red-team capabilities for defense research.

Benchmark Performance

Independent evaluations · Artificial Analysis

46.4%
Intelligence

Accuracy & Capability Details

Humanity's Last Exam43.1%
SciCode - Scientific Coding57.4%
Long Context Reasoning76.7%

Compare Models Side-by-Side

Evaluate specifications, pricing, and independent benchmark indices

Model Details
General Info
ProviderSpaceXAIAnthropicAnthropic
Release DateSeptember 21, 2026September 22, 2026September 28, 2026
Knowledge Cutoff--Jun 2026
Context & Limits
Context Window500K
1M
Best Context Window
1M
Best Context Window
Pricing (per 1M tokens)
Input Pricing
$2
Best Input Pricing
$4
$2
Best Input Pricing
Output Pricing
$6
Best Output Pricing
$20$10
Modalities
Inputs
textimagefile
textimagefile
textimagefile
Outputs
text
text
text
Benchmarks (0-100)
Intelligence Index46.4
57.6
Best Intelligence Index
56.0
Coding Index---
Agentic Index---
Grok 4.7
Claude Opus 5.5
Claude Sonnet 5.5

Humanity's Last Exam

Extremely difficult logical reasoning and knowledge.

43%
Grok 4.7
Humanity's Last Exam
Score: 43%
Grok 4.7
61%
Claude Opus 5.5
Humanity's Last Exam
Score: 61%
Claude Opus 5.5
55%
Claude Sonnet 5.5
Humanity's Last Exam
Score: 55%
Claude Sonnet 5.5

Long Context Reasoning

Logical reasoning over long context windows.

77%
Grok 4.7
Long Context Reasoning
Score: 77%
Grok 4.7
85%
Claude Opus 5.5
Long Context Reasoning
Score: 85%
Claude Opus 5.5
83%
Claude Sonnet 5.5
Long Context Reasoning
Score: 83%
Claude Sonnet 5.5

SciCode Benchmark

Scientific coding and mathematical modeling.

57%
Grok 4.7
SciCode Benchmark
Score: 57%
Grok 4.7
67%
Claude Opus 5.5
SciCode Benchmark
Score: 67%
Claude Opus 5.5
61%
Claude Sonnet 5.5
SciCode Benchmark
Score: 61%
Claude Sonnet 5.5

Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.

Explore more from SpaceXAI

Other models by SpaceXAI

Top AI Models

Leading alternatives by intelligence score

View all