vram.run Models Hardware Providers Cloud State of Inference
API provider data is live · Hardware & cloud pricing curated 2026-02-23

Kimi-K2-Thinking

by moonshotai

1058B params · text-generation · 1.7k likes · 151.9k downloads

Kimi-K2-Thinking is a 1058B parameter model. At Q4 quantization it requires 529GB of VRAM. It requires a GPU with at least 529GB of VRAM.

Inference providers

Provider$/1M in$/1M outThroughput
Novita62 tok/s
Featherless
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0