Grok 4.20 (Non-Reasoning)

Grok 4.20 (Non-Reasoning) is a model from xAI for agentic tool use, reasoning, coding, and live assistance. It is the first version of the grok line in the registry. It has a 1,000,000 token context window and a 30,000 token output limit. Pricing starts at $1.25 per million input tokens and $2.50 per million output tokens, with cached input at $0.20 per million. Weights are not open. It accepts image, pdf, and text, and returns text. It supports tool calling but no reasoning mode.

First version of its line in the registry.

Range across the hosts that serve it: up to $2.00 per 1M input tokens.

• ·Agentic tool use for live assistance
• ·1M token context window
• ·30K token output limit
• ·Accepts image, pdf, and text
• ·Tool calling supported

• Reach for it for agentic tool use tasks
• Reach for it for coding assistance
• Reach for it for live assistance scenarios
• Reach for it for processing images, PDFs, and text

6 hosts serve this model at the price above · the maker's documentation

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window size?

1,000,000 tokens.

Does it support reasoning mode?

No, it is a non-reasoning model.

What is the price per million tokens?

Input from $1.25, output from $2.50, cached input from $0.20.