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

DeepSeek-V4-Pro

by deepseek-ai

862B params · text-generation · 5.0k likes · 2.6M downloads

DeepSeek-V4-Pro is a 862B parameter model. At Q4 quantization it requires 431GB of VRAM. It requires a GPU with at least 431GB of VRAM.

Inference providers

Provider$/1M in$/1M outThroughput
Novita59 tok/s
Together AI57 tok/s
Fireworks47 tok/s
Featherless
deepinfra40 tok/s
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0