Compare AI models

Specifications
Parameters
236B48B
Context window
128k1M
These models have no shared benchmark scores.
Overview
CompanyDeepSeekMoonshot AI
Release dateJun 28 2024Oct 30 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 and Kimi Linear don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. DeepSeek V2-0628 shipped 489 days before Kimi Linear, so benchmark comparisons should account for the intervening progress.

DeepSeek V2-0628 has 236B parameters, while Kimi Linear has 48B. Context windows are 128k (DeepSeek V2-0628) vs 1M (Kimi Linear).

Direct benchmark comparisons are unavailable — DeepSeek V2-0628 and Kimi Linear don't publish scores on any of the same benchmarks.