deepseek-ai/DeepSeek-V3.1

Text generation · served with vLLM · ↓ 340.2K downloads a month · View on Hugging Face

Weights: 689 GB
GPU memory needed: about 783.2 GB
Conversation length: 8,192 tokens
Model allows: 163,840 tokens
Memory · disk: 352.3 GB · 798 GB
Architecture: DeepseekV3ForCausalLM
Licence: mit

Versions of DeepSeek ~685B

The same model packaged differently: a quantized version needs less GPU memory and runs on a cheaper machine.

Version	Runtime	Weights	GPU memory	Cheapest machine now
Originaldeepseek-ai/DeepSeek-V3.1-Base	vLLM	689 GB	783.2 GB	$44.03/h 8x Nvidia H200
Maindeepseek-ai/DeepSeek-V3.1	vLLM	689 GB	783.2 GB	$44.03/h 8x Nvidia H200
Terminusdeepseek-ai/DeepSeek-V3.1-Terminus	vLLM	689 GB	783.2 GB	$44.03/h 8x Nvidia H200

Hardware and deploy

Endpoint name

$80.35/h from your balance while it runs, billed by the hour; stop it from its page. GridPup rents the machine with vLLM, which downloads the model from Hugging Face (usually ready in 10 to 15 minutes), and gives you an https address, keys and a Playground.

Loading…