Compare AI models

Specifications
Parameters
1.5B/7B/8B/14B/32B/70B—
Context window
—128k
Benchmarks
LiveCodeBench
57.5%62.5%
MATH-500
94.5%96.2%
AIME 2024
70%77.5%
Overview
CompanyDeepSeekMoonshot AI
Release dateJan 20 2025Jan 20 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Kimi K1.5 leads DeepSeek R1 Distill on 3 of the 3 benchmarks they both report (LiveCodeBench, MATH-500, AIME 2024). Both models were released on the same day — Jan 20 2025.

DeepSeek R1 Distill is open weight, while Kimi K1.5 is closed.

On LiveCodeBench, Kimi K1.5 leads at 62.5% vs DeepSeek R1 Distill at 57.5%. On MATH-500, Kimi K1.5 leads at 96.2% vs DeepSeek R1 Distill at 94.5%. On AIME 2024, Kimi K1.5 leads at 77.5% vs DeepSeek R1 Distill at 70%.