X

xAI: grok-4.20-non-reasoning:free

grok-4.20-non-reasoning:free
1M contextToolsVisionFilesCacheStructuredWeb search
Released Mar 31, 2026Knowledge cutoff Sep 2025Updated Jun 20, 2026

Grok-4.20-beta-0309-non-reasoning is the ultra-fast, efficiency-first variant of the Grok-4.20 family. Released as part of the March 2026 update, this model bypasses the "Reasoning Trace" (Chain-of-Thought) layers to provide near-instantaneous responses for high-volume data processing. It is powered by the Colossus supercluster and features a "Rapid Learning" architecture that incorporates real-time X platform data with millisecond latency. This model is ideal for tasks where immediate pattern recognition, fast drafting, or real-time information retrieval from X (Twitter) is required, offering the lowest hallucination rate in the non-reasoning class through its "Society of Mind" multi-agent verification (performed silently in the background).

Mode chatTokenizer Grok

Pricing

Input price
$0.00/ 1M tokens
Output price
$0.00/ 1M tokens
Compatible endpoints openaiVendor xAI

Uptime

Performance

Loading performance data...

Usage & Ranking

Loading usage...

Supported parameters

All providers = every upstream serving this model supports it. Some providers = depends on which upstream handles the request. Default = the value sent when you leave the parameter unset.

ParameterProvidersDefault
frequency_penaltySome providersNot sent by default
include_reasoningAll providers-
logprobsAll providers-
max_tokensAll providers-
presence_penaltySome providersNot sent by default
reasoningAll providers-
repetition_penaltySome providersNot sent by default
response_formatAll providers-
seedAll providers-
structured_outputsAll providers-
temperatureAll providersNot sent by default
tool_choiceAll providers-
toolsAll providers-
top_kSome providersNot sent by default
top_logprobsAll providers-
top_pAll providersNot sent by default

Frequently asked questions

Similar models