RTX 5090 32GB Class Cloud GPU Price Comparison — RunPod vs Vast.ai Live Rates

Live on-demand rates for RTX 5090 32GB Class cards from RunPod Secure, RunPod Community and Vast.ai Verified, plus a 1x/2x compatibility matrix for 16 open-weight models at 32K context.

Live rates — RunPod Secure vs RunPod Community vs Vast.ai Verified

GPUVRAM (GB)RunPod Secure ($/hr)RunPod Community ($/hr)Vast.ai Verified ($/hr)Lowest ($/hr)Vast live rigsRunPod Secure 730hVast 730hDeploy
RTX 5090 32GB32$1.39$0.69$0.50$0.504$1014.70$362.81Vast.aiRunPod
A100 SXM 40GB40—$1.00—$1.000——RunPodVast.ai
RTX A6000 48GB48$0.69$0.33$0.40$0.335$503.70$293.61RunPodVast.ai

Dual/quad reference for RTX 5090 32GB: RunPod 2x from $1.38/hr, Vast.ai 2x from $0.99/hr, lowest 2x floor $0.99/hr. 4x floors: RunPod $2.76, Vast.ai $1.99, lowest $1.99/hr.

Open-model compatibility — 1x vs 2x at 32K context

Open model (32K ctx)RTX 5090 32GB 1xRTX 5090 32GB 2xA100 SXM 40GB 1xA100 SXM 40GB 2xRTX A6000 48GB 1xRTX A6000 48GB 2x
Qwen3.8-27BRequires 2x+
Requires 2x+
Fits INT4 (32K)
Fits FP16 (32K)
Fits FP8 (32K)
Fits INT4 (32K)
Requires 2x+
Fits FP8 (32K)
Fits INT4 (32K)
Fits FP16 (32K)
Fits FP8 (32K)
Fits INT4 (32K)
Requires 2x+
Fits FP8 (32K)
Fits INT4 (32K)
Fits FP16 (32K)
Fits FP8 (32K)
Fits INT4 (32K)
Qwen3.8-Flash-NextRequires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Arete-27BRequires 2x+
Requires 2x+
Fits INT4 (32K)
Fits FP16 (32K)
Fits FP8 (32K)
Fits INT4 (32K)
Requires 2x+
Fits FP8 (32K)
Fits INT4 (32K)
Fits FP16 (32K)
Fits FP8 (32K)
Fits INT4 (32K)
Requires 2x+
Fits FP8 (32K)
Fits INT4 (32K)
Fits FP16 (32K)
Fits FP8 (32K)
Fits INT4 (32K)
diffusiongemma-26B-A4B-itRequires 2x+
Requires 2x+
Fits INT4 (32K)
Fits FP16 (32K)
Fits FP8 (32K)
Fits INT4 (32K)
Requires 2x+
Fits FP8 (32K)
Fits INT4 (32K)
Fits FP16 (32K)
Fits FP8 (32K)
Fits INT4 (32K)
Requires 2x+
Fits FP8 (32K)
Fits INT4 (32K)
Fits FP16 (32K)
Fits FP8 (32K)
Fits INT4 (32K)
MiniMax-M3.1-FlashRequires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
GLM-5.3-FlashRequires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
GLM-5.3Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
GLM-5.2Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
DeepSeek-V4.1-Flash-1017Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
DeepSeek-V4.1-FlashRequires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
DeepSeek-V4-Flash-0731Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
DeepSeek-V4-Pro-0813Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Kimi-K2.7-CodeRequires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Kimi-K3.1-ProRequires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Qwen3.8-2.4T-A95BRequires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
DeepSeek-V4-Flash-DSparkRequires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x
Requires 2x+
Requires 2x+
Requires 2x+
No fit at 2x
No fit at 2x
No fit at 2x

Verdicts hard-computed from each repo's safetensors and config: FP16 (16-bit), FP8 (8-bit) and INT4 (4.5-bit effective) weights + GQA KV cache at 32768 tokens + runtime overhead, concurrency 1.

What the monthly gap looks like

Rent a RTX 5090 32GB around the clock (730 h/month): RunPod Secure bills $1014.70/mo, while the cheapest verified, rentable Vast.ai bundle bills $362.81/mo — a real gap of $651.89 every month ($7822.68 per year) for the identical GPU class. Locking a yearly hyperscaler contract instead of checking this table is where the real money leaks.