NVIDIA GeForce RTX 3090

NVIDIA24GB GDDR6XPCIe Gen4Ampere

Data updated 2026-11-04 15:09 UTC · refreshed every 5 min

Current Price

$0.37/hr

From $0.15/hr · lowest live offer

320 GPU · High availability

Current Price

$0.37

+20.5% (24h)

Available Now

320 GPU

High availability

Total in Network

993 GPU

Live data

Utilization

67.8%

Stable

Market history

Price Tracking

USD/hour (minimum · median)

Available GPUs

Ready to rent

Total in Network

Infrastructure Growth

Which LLMs run on one NVIDIA GeForce RTX 3090

Open VRAM calculator with RTX 3090

Estimated with our VRAM calculator for a single card: 8K context, 4 concurrent users, vLLM-style serving.

Largest that fitsINT4 / AWQ

GPT-OSS 20B

21B MoE · 3.6B active

14.2 of 24 GB59%

~75 tok/s per stream

Details

Where the memory goes

  • Model weights12.24 GB
  • KV cache0.82 GB
  • Activations0.02 GB
  • CUDA & framework1.12 GB
Precision
INT4 / AWQ
Serving preset
8K context × 4 users
Users at this context
51
Spare context in leftover VRAM
329K
Fit score
79/100 · Moderate
Open in VRAM calculator
BalancedFP8

Qwen3.5-9B

9B

14.9 of 24 GB62%

~54 tok/s per stream

Details

Where the memory goes

  • Model weights12.69 GB
  • KV cache1.07 GB
  • Activations0.05 GB
  • CUDA & framework1.05 GB
Precision
FP8
Serving preset
8K context × 4 users
Users at this context
38
Spare context in leftover VRAM
239K
Fit score
77/100 · Moderate
Open in VRAM calculator
CompactFP16

Qwen3.5-4B

4B

11.4 of 24 GB47%

~69 tok/s per stream

Details

Where the memory goes

  • Model weights9.20 GB
  • KV cache1.07 GB
  • Activations0.06 GB
  • CUDA & framework1.02 GB
Precision
FP16
Serving preset
8K context × 4 users
Users at this context
51
Spare context in leftover VRAM
345K
Fit score
86/100 · Moderate
Open in VRAM calculator

Estimates, not benchmarks: real usage depends on the runtime, context and batch settings.

Technical Specifications

Memory

24GB GDDR6X

FP32 Performance

35.6 TFLOPS

Tensor Performance

285 TFLOPS

Memory Bandwidth

936 GB/s

Architecture

Ampere

TDP

350W

FP64 Performance

0.56 TFLOPS

CUDA Cores

10496

Tensor Cores

328

Ideal Use Cases

Machine LearningDeep Learning3D RenderingGamingVideo Production

About NVIDIA GeForce RTX 3090

The RTX 3090 remains a capable option for AI workloads with its 24GB of memory.

The NVIDIA GeForce RTX 3090 features 24GB of GDDR6X memory with 936 GB/s bandwidth, making it excellent for machine learning, deep learning, 3d rendering. With 285 TFLOPS tensor performance and 35.6 TFLOPS FP32 compute, this NVIDIA Ampere GPU delivers exceptional performance for demanding workloads.

Currently, 993 GPU RTX 3090 GPUs are available in the GPUniq network with 67.8% utilization. The average rental price is $0.37 per hour, with 320 GPU ready for immediate deployment.

RTX 3090 vs similar GPUs — specs & live rental prices

Prices are live minimum / median USD per hour on the GPUniq marketplace, as of 2026-11-04 15:09 UTC.

GPUVRAMArchitectureFP32TensorBandwidthTDPMin $/hrMedian $/hrAvailable
RTX 309024GB GDDR6XAmpere35.6 TFLOPS285 TFLOPS936 GB/s350W$0.16$0.40285
RTX 3090 Ti24GB GDDR6XAmpere40 TFLOPS320 TFLOPS1.01 TB/s450W$0.18$0.3614
RTX 3080 Ti12GB GDDR6XAmpere34.1 TFLOPS273 TFLOPS912 GB/s350W$0.15$0.30183
RTX 308010GB GDDR6XAmpere29.8 TFLOPS238 TFLOPS760 GB/s320W$0.13$0.30552

Frequently Asked Questions

How much does it cost to rent NVIDIA GeForce RTX 3090?

The current rental price for NVIDIA GeForce RTX 3090 is $0.37 per hour on GPUniq. Prices may vary based on demand and availability.

What is NVIDIA GeForce RTX 3090 best used for?

NVIDIA GeForce RTX 3090 is ideal for Machine Learning, Deep Learning, 3D Rendering, Gaming, Video Production. With 24GB of GDDR6X memory and 285 TFLOPS tensor performance, it excels at demanding AI and compute workloads.

How many NVIDIA GeForce RTX 3090 GPUs are available for rent?

Currently, there are 320 GPU available for immediate rental out of 993 GPU total in the GPUniq network. The current utilization is 67.8%.

Which LLMs can run on NVIDIA GeForce RTX 3090?

On a single NVIDIA GeForce RTX 3090, our VRAM calculator fits GPT-OSS 20B (INT4 / AWQ, 14.2 of 24 GB); Qwen3.5-9B (FP8, 14.9 of 24 GB); Qwen3.5-4B (FP16, 11.4 of 24 GB) — estimated at 8K context with 4 concurrent users.

What are the specifications of NVIDIA GeForce RTX 3090?

NVIDIA GeForce RTX 3090 features 24GB GDDR6X memory, 35.6 TFLOPS FP32 performance, 285 TFLOPS tensor performance, 936 GB/s memory bandwidth, and 350W TDP. It uses the Ampere architecture.

Related GPUs in GeForce RTX 30 Series

GPU rental guides

NVIDIA official specificationsAPI docsAbout GPUniqContact supportPrivacy policyTerms of use

$0.37/hr

RTX 3090 · 320 GPU

Rent now