Compare AI models

Specifications
Parameters
1T744B
Context window
256k—
API pricing
Input price
$0.60$1.00
Output price
$3.00$3.20
Cached input price
$0.10$0.20
Cheapest input
$0.45SiliconFlow$0.60GMICloud
Cheapest output
$2.25SiliconFlow$1.92GMICloud
Benchmarks
BullshitBench v2
52%28%
SWE-Bench Verified
76.8%77.8%
BrowseComp
60.6%75.9%
Humanity's Last Exam · with tools
30.1%50.4%
GPQA Diamond
87.6%86%
Overview
CompanyMoonshot AIZ.ai
Release dateJan 27 2026Feb 12 2026
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

GLM-5 leads Kimi K2.5 on 3 of the 5 benchmarks they both report. Kimi K2.5 is cheaper on both input and output: $0.60 vs $1.00 per million input tokens, and $3.00 vs $3.20 per million output tokens. Kimi K2.5 shipped 16 days before GLM-5, so benchmark comparisons should account for the intervening progress.

Kimi K2.5 has 1T parameters, while GLM-5 has 744B.

On BullshitBench v2, Kimi K2.5 leads at 52% vs GLM-5 at 28%. On SWE-Bench Verified, GLM-5 leads at 77.8% vs Kimi K2.5 at 76.8%. On BrowseComp, GLM-5 leads at 75.9% vs Kimi K2.5 at 60.6%. On Humanity's Last Exam · with tools, GLM-5 leads at 50.4% vs Kimi K2.5 at 30.1%. On GPQA Diamond, Kimi K2.5 leads at 87.6% vs GLM-5 at 86%.