Skip to main content

Grok 4.1 Fast

xAI

๐Ÿง  Reasoning๐ŸŒ Web Search

A budget speedster with a huge 2M-token context window. One of the cheapest ways to process large amounts of text.

Released November 19, 2025

Pricing

Input tokens$0.20/M
Output tokens$0.50/M

Capacity

Context window2.0M tokens

Capabilities

โœ“Reasoning & Planning
โœ“Web Search
โœ—Open Source

Best Scores

GPQA Diamond85.3%
SWE-bench Pro70.0%
Humanity's Last Exam17.6%

Best For

๐Ÿ“summarization
๐Ÿ“Šanalysis
๐Ÿ’ปcoding

Benchmark Scores

Knowledge

GPQA Diamond

PhD-level science questions written so you cannot just Google the answer. Tests whether the model can reason about hard science.

85.3%

Independently measured by Artificial Analysis.

Humanity's Last Exam

PhD-level questions across many subjects. Tests deep reasoning on the hardest questions humans can ask.

17.6%

Independently measured by Artificial Analysis.

Software Engineering

SWE-bench Pro

The harder version of the coding test. Bigger codebases, trickier bugs. Scores drop for everyone, so the gaps between models become clearer.

70.0%

Provider-reported; no independent run recorded yet.

Why Choose Grok 4.1 Fast?

Grok 4.1 Fast is an absolute steal for bulk text processing. At $0.20/$0.50, it's the cheapest frontier model, and the 2M context is unmatched. Perfect for summarizing books, processing logs, or handling large datasets on a budget.

How It Compares

vs Claude Haiku 4.5

Grok Fast is 5x cheaper with 10x more context; Haiku is more capable.

vs Llama 4 Scout

Grok Fast is faster and more capable; Scout has slightly more context.

You Might Also Like