Source URL: https://simonwillison.net/2025/Sep/20/grok-4-fast/
Source: Simon Willison’s Weblog
Title: Grok 4 Fast
Feedly Summary: Grok 4 Fast
New hosted reasoning model from xAI that’s designed to be fast and extremely competitive on price. It has a 2 million token context window and “was trained end-to-end with tool-use reinforcement learning".
It’s priced at $0.20/million input tokens and $0.50/million output tokens – 15x less than Grok 4 (which is $3/million input and $15/million output). That puts it cheaper than GPT-5 mini and Gemini 2.5 Flash on llm-prices.com.
The same model weights handle reasoning and non-reasoning based on a parameter passed to the model.
I’ve been trying it out via my updated llm-openrouter plugin, since Grok 4 Fast is available for free on OpenRouter for a limited period.
The non-reasoning model:
llm -m openrouter/x-ai/grok-4-fast:free "Generate an SVG of a pelican riding a bicycle"
And the reasoning model:
llm -m openrouter/x-ai/grok-4-fast:free "Generate an SVG of a pelican riding a bicycle" -o reasoning_enabled true
In related news, the New York Times had a story a couple of days ago about Elon’s recent focus on xAI: Since Leaving Washington, Elon Musk Has Been All In on His A.I. Company.
Tags: ai, generative-ai, llms, llm, llm-pricing, pelican-riding-a-bicycle, llm-reasoning, grok, llm-release, openrouter
AI Summary and Description: Yes
Summary: The text introduces Grok 4 Fast, a new hosted reasoning model from xAI, highlighting its competitive pricing and capabilities. Professionals in AI and cloud security should note the model’s unique features and its impact on pricing trends in the AI landscape.
Detailed Description: The release of Grok 4 Fast is significant within the AI sector, particularly for those interested in large language models (LLMs). Here are the key points and implications for security and compliance professionals:
– **Model Overview**: Grok 4 Fast is a hosted reasoning model designed to be highly efficient and competitively priced.
– **Technical Specifications**:
– Supports a **2 million token context window**, allowing for substantial input capacity.
– Uses an **end-to-end tool-use reinforcement learning** approach for training, which can enhance model efficacy in specific applications.
– **Pricing Structure**:
– Priced at **$0.20/million input tokens** and **$0.50/million output tokens**. This price point is significantly lower than its predecessor, Grok 4, and competitive with other models like GPT-5 mini and Gemini 2.5 Flash.
– Such pricing could encourage broader adoption of reasoning models, potentially democratizing access to advanced AI capabilities.
– **Use-Cases**:
– The model can handle both reasoning and non-reasoning tasks based on parameters set by users, indicating its versatility.
– Example commands illustrate its practical applications, such as generating SVG artwork with specified reasoning.
– **Market Context**:
– The release comes amid increased interest in AI technologies and raises considerations about xAI’s strategic direction under Elon Musk’s vision, as noted in recent media.
The combination of competitive pricing, advanced capabilities, and the context of prominent figures in AI makes Grok 4 Fast a relevant development for professionals in AI, security, and compliance fields. Monitoring such advancements is crucial as they could influence data governance, compliance strategies, and the overall landscape of AI technologies.