Compare AI models

Specifications
Parameters
1T—
Context window
256k—
API pricing
Cheapest input
$0.49AtlasCloud—
Cheapest output
$2.50AtlasCloud—
$223AtlasCloud · int4—
Benchmarks
Humanity's Last Exam · with tools
30.1%44.4%
Overview
CompanyMoonshot AISpaceXSI
Release dateJan 27 2026Jul 9 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Grok 4 Heavy leads Kimi K2.5 on 1 of the 1 benchmark they both report (Humanity's Last Exam). Grok 4 Heavy shipped 202 days before Kimi K2.5, so benchmark comparisons should account for the intervening progress.

Kimi K2.5 is open weight, while Grok 4 Heavy is closed.

On Humanity's Last Exam · with tools, Grok 4 Heavy leads at 44.4% vs Kimi K2.5 at 30.1%.