Model provider cost

Host DeepSeek-R1 Distill Qwen 32B on Vast.ai

DeepSeek-R1 Distill Qwen 32B needs 1x 48GB+ GPU, and Vast.ai's current cheapest qualifying row is L40 at $0.45/hr.

1x 48GB+ GPU 32B params High-end reasoning MIT
Cheapest on this provider
$0.45/hr
L40
Monthly estimate
$331/mo
730 hours at the current median
VRAM baseline
48GB
1x 48GB+ GPU
Qualifying rows
8
Updated Jun 21, 2026

Vast.ai rows that can host DeepSeek-R1 Distill Qwen 32B

The cheapest tracked way to host DeepSeek-R1 Distill Qwen 32B on Vast.ai is L40 at $0.45/hr. Vast.ai is currently tied for the cheapest tracked market setup for this model.

GPU VRAM Per GPU Estimated hourly Estimated monthly Updated
L40 48GB $0.45/hr $0.45/hr $331/mo Jun 21, 2026
RTX 6000Ada 48GB $0.60/hr $0.60/hr $439/mo Jun 21, 2026
A100 PCIE 80GB $1.00/hr $1.00/hr $731/mo Jun 19, 2026
A100 SXM4 80GB $1.14/hr $1.14/hr $834/mo Jun 21, 2026
H100 PCIE 80GB $1.64/hr $1.64/hr $1,199/mo Jun 21, 2026
H100 NVL 94GB $2.27/hr $2.27/hr $1,656/mo Jun 21, 2026
H100 SXM 80GB $2.40/hr $2.40/hr $1,752/mo Jun 21, 2026
H200 NVL 141GB $3.46/hr $3.46/hr $2,529/mo Jun 21, 2026

Why this setup does or does not fit

VRAM floor

1x 48GB+ GPU

1x 48GB GPU or better. Long prompts, batching, and KV cache can require extra headroom.

Model quality

High-end reasoning

Sharper planning and analysis than small models, with a corresponding jump in latency and memory needs.

Operational note

DeepSeek 32B

A good reasoning upgrade when you want chain-of-thought style behavior without a cluster.

DeepSeek-R1 Distill Qwen 32B on Vast.ai FAQ

Can I host DeepSeek-R1 Distill Qwen 32B on Vast.ai?

The cheapest tracked way to host DeepSeek-R1 Distill Qwen 32B on Vast.ai is L40 at $0.45/hr.

What GPU memory does DeepSeek-R1 Distill Qwen 32B need?

Our baseline for DeepSeek-R1 Distill Qwen 32B is 1x 48GB+ GPU. The practical recommendation is 1x 48GB GPU or better.

Is Vast.ai the cheapest provider for DeepSeek-R1 Distill Qwen 32B?

Vast.ai is currently tied for the cheapest tracked market setup for this model.

How fresh is this Vast.ai DeepSeek-R1 Distill Qwen 32B cost page?

This page recalculates from the latest tracked on-demand rows. The freshest qualifying Vast.ai row shown here is from Jun 21, 2026.

Compare this setup