AI Hub
← All models

GLM 4.7 vs Kimi K2-Instruct-0905

Zhipu AI vs Moonshot AI — benchmarks, pricing, and capabilities side by side.

  • GLM 4.7 has the higher intelligence index (89 vs 66)
GLM 4.7Kimi K2-Instruct-0905
Intelligence index8966
DeveloperZhipu AIMoonshot AI
TypeLLMLLM
AccessOpen weightsOpen weights
Context window202,752 tokens
Input price$0.40 / 1M
Output price$1.75 / 1M
Speed98 tok/s
ReleasedDecember 22, 2025September 5, 2025
Parameters1000000000000
Input modalitiesText
Output modalitiesText

Shared benchmarks

GLM 4.7
Kimi K2-Instruct-0905
AIME 2025
95
49.5
GPQA Diamond
85.9
75.1
Humanity’s Last Exam
25.1
4.7
LiveCodeBench
89.4
53.7
MMLU-Pro
85.6
81.1