← All models
Compare

GLM-5.2 vs Kimi K3

Add, remove, or swap models to compare them side by side.

GLM-5.2Kimi K3
AttributeGLM-5.2Z.ai (Zhipu AI) · 17 in the newsKimi K3Moonshot · 22 in the news
Scores
Intelligence (ECI)152157
Coding8599
Math3583
Reasoning & Knowledge6279
Agentic & Tools9086
Specifications
DeveloperZ.ai (Zhipu AI)Moonshot
FamilyKimi
ReleasedJun 16, 2026Jul 16, 2026
Parameters2.8T
AvailabilityOpen weights (unrestricted)Open weights (non-commercial)
Context window1M1M
Price — $/M input$0.97$3.00
Price — $/M output$3.04$14.00
Inputstexttext, image
Outputstexttext
Benchmarks
AIME 2024/202586%97%
APEX36%39%
ARC-AGI77%95%
ARC-AGI-223%60%
GPQA Diamond92%93%
Humanity's Last Exam40%44%
SciCode51%59%
SimpleQA Verified38%43%
SWE-bench Verified79%
WebDev Arena15931682
WeirdML70%83%
τ²-bench99%

Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.

Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.