Grok 4.20
All models
xAIxAIGrokReleased 2026-03-31

Grok 4.20

2M context$2.00/M input$6.00/M output

Grok 4.20 is xAI's flagship model, notable for its industry-leading speed and agentic tool-calling capabilities. It features a 2 million token context window and delivers precise and truthful responses with a low hallucination rate and strict prompt adherence. The model supports reasoning, which can be enabled or disabled via API parameters, allowing for step-by-step thinking processes.

What is Grok 4.20?

Grok 4.20 is an AI model from xAI that Agent Mag tracks for pricing, context window, modalities, benchmarks, and API compatibility. Builders can use this page to compare Grok 4.20 against other models for agent workflows and production deployments.

Model ID

Grok 4.20 is xAI's flagship model, notable for its industry-leading speed and agentic tool-calling capabilities. It features a 2 million token context window and delivers precise and truthful responses with a low hallucination rate and strict prompt adherence. The model supports reasoning, which can be enabled or disabled via API parameters, allowing for step-by-step thinking processes.

Architecture & Specifications
Tokenizer
Grok
Training Data
Knowledge cutoff September 1, 2025
License
Proprietary
Released
2026-03-31
Modalities
Input
textimagefile
Output
text
Supported Parameters
include_reasoninglogprobsmax_tokensreasoningresponse_formatseedstructured_outputstemperaturetool_choicetoolstop_logprobstop_p
Strengths
  • Industry-leading speed
  • Agentic tool-calling capabilities
  • Low hallucination rate
  • Strict prompt adherence
  • Supports reasoning with step-by-step thinking
Limitations
  • Requires user IDs for anonymity
  • Moderation is the responsibility of the developer
Recommended Use Cases
Roleplay
Marketing
Agentic tool usage
Structured outputs
Deep reasoning tasks

Related content

Data enriched Apr 24, 2026. Pricing from OpenRouter API.