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

DeepSeek-V3.1

by deepseek-ai

685B params · text-generation · 818 likes · 149.8k downloads

DeepSeek-V3.1 is a 685B parameter model. At Q4 quantization it requires 342GB of VRAM. It requires a GPU with at least 342GB of VRAM.

Inference providers

Provider$/1M in$/1M outThroughput
Novita36 tok/s
Together AI40 tok/s
Fireworks79 tok/s
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0