DeepSeek logo

Rent DeepSeek-R1 7B in a private loop

Compact reasoning model
By DeepSeekdeepseek-r1:7b7B parameters32K tokens of contextMIT

The cheapest way to try reasoning: the same think-out-loud manner as bigger R1 models, but it fits a single inexpensive card. Good at school and olympiad maths, weaker on general knowledge.

GPU configurations for DeepSeek-R1 7B

1x V100

Best value
~28 tok/sup to 2 streams
$0.22per hour

1x RTX4090

~82 tok/sup to 4 streams
$0.56per hour

1x RTX5090

~108 tok/sup to 6 streams
$0.65per hour

1x A100

~72 tok/sup to 8 streams
$1.45per hour

1x H100

~118 tok/sup to 16 streams
$3.33per hour
Pick a GPU
VRAM
RAM
Storage
vCPU
CPU

Billing follows server uptime by the hour.

DeepSeek-R1 7B benchmarks

as published by the model vendor
DeepSeek-R1 7B scores in public benchmarks
MATH-50092.8
AIME 202455.5
GPQA Diamond49.1

Other models in the private loop

FAQ

Frequently asked about renting DeepSeek-R1 7B

Which GPU does DeepSeek-R1 7B need?

The entry configuration is 1x V100, which delivers around 28 tokens per second. For more parallel requests and headroom on speed, go with 1x H100.

How much does DeepSeek-R1 7B cost?

An hour of DeepSeek-R1 7B runs from $0.22 to $3.33 depending on the GPU configuration. Billing is hourly: you pay for server time, not for tokens.

How do I call DeepSeek-R1 7B from my code?

Once deployed you get an OpenAI-compatible endpoint: point any OpenAI client at the base_url you receive and pass deepseek-r1:7b as the model name. Code written against the OpenAI API needs no changes.

Does the data stay private?

Yes. DeepSeek-R1 7B runs on a dedicated GPU server inside a private loop: requests and responses never reach the model vendor and are not used for training.

What context window and license does DeepSeek-R1 7B have?

DeepSeek-R1 7B holds 32K tokens of context per request and ships under the MIT license, so you can use it commercially on your own hosting terms.