Grok 4.20 0309 (Non-reasoning)

Last updated: Recently

Model Type

Agent

Max Context

2000K

Max Output

Parameter Scale

Grok 4.20 is reasoning model from xAI with industry leading speed and agentic tool calling capabilities. 2,000,000 token context window.

Input Modality

text, image, file

Output Modality

text

Inference Speed

81.617 tokens/s

Latest Release Date

3/10/2026

SDK Ecosystem

artificial-analysismanufactureraa-bootstrap

Model Overview Fields

name

Grok 4.20 0309 (Non-reasoning)

Release date

3/10/2026

Performance

81.617 tokens/s

model_creator.name

xAI

AA Evaluation Scores

AA Intelligence Index

29.7

AA Coding Index

25.4

AA Math Index

MMLU Pro

GPQA

0.785

HLE

0.225

LiveCodeBench

SciCode

0.322

Math-500

AIME

AIME 2025

IFBench

0.478231292517007

LCR

0.18

TerminalBench Hard

0.21969696969697

TAU2

0.695906432748538

Output Speed (tokens/s)

81.617 tok/s

TTFT (s)

0.716s

First Answer Token (s)

0.716s

xAI

Grok 4.20 0309 (Non-reasoning)

$2.00/M$6.00/M

Grok 4.20 is reasoning model from xAI with industry leading speed and agentic tool calling capabilities. 2,000,000 token context window.

There are no reviews for this model yet.

artificial-analysismanufactureraa-bootstrap

0.0