xpu liveBETA
GPU comparisonGPU providers

GPU CLOUD DIRECTORY

Find your next GPU.

Compare hardware and rental prices.

  • 84 GPUs
  • 5 providers
  • 83 offers
Meet the providers ↗

HARDWARE & PRICING

GPU explorer 84

How prices compare ↗

84 GPUs

Scroll horizontally to compare all specifications and prices ↔

GPU specifications and published rental prices in US dollars per GPU-hour
GPU modelVRAMArchitectureWorkloadsProvidersListed from USD / GPU-hour
AMD Instinct MI300X — MI300X reference product view. Source: AMD.AMD Instinct MI300XAMD · OAM module↗ 192 GBHBM3 CDNA 3 LLM inference, training and HPC Runpod $2.39Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
AMD Instinct MI350X — MI350X reference product view. Source: AMD.AMD Instinct MI350XAMD · OAM module↗ 288 GBHBM3e CDNA 4 LLM inference, training and HPC CloudRiftRunpod $5.49Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA A10 — Reference hardware; rental provider cooling configuration may differ.A10NVIDIA · PCIe↗ 24 GBGDDR6 Ampere Inference and graphics ModalVast.ai $0.2289Updated 2026-11-22 · 15:43 UTCOn-demand · Vast.aiPublished list price
NVIDIA A100 PCIe 80GB — Reference hardware; rental provider cooling configuration may differ.A100 PCIe 80GBNVIDIA · PCIe↗ 80 GBHBM2e Ampere Training and inference Runpod $1.19Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA A100 SXM 40GB — NVIDIA A100 SXM family reference image; this page and rental identity are the 40 GB version.A100 SXM 40GBNVIDIA · SXM4↗ 40 GBHBM2 Ampere LLM inference, training and HPC Runpod $1.00Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA A100 SXM 80GB — Reference hardware; rental provider cooling configuration may differ.A100 SXM 80GBNVIDIA · SXM4↗ 80 GBHBM2e Ampere Training and fine-tuning CloudRiftRunpodVast.ai $1.0570Updated 2026-11-22 · 15:43 UTCOn-demand · Vast.aiPublished list price
NVIDIA A40 — Reference hardware; rental provider cooling configuration may differ.A40NVIDIA · PCIe↗ 48 GBGDDR6 Ampere Visualization and AI workloads RunpodVast.ai $0.5489Updated 2026-11-22 · 15:43 UTCOn-demand · Vast.aiPublished list price
NVIDIA B200 — NVIDIA B200 SXM6 single GPU module with heatsink; photo: ServeTheHome / Patrick Kennedy.B200NVIDIA · SXM↗ 180 GBHBM3e Blackwell Large-model training and inference ModalRunpodVast.ai $5.98Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA B300 — B300 reference product view. Source: NVIDIA.B300NVIDIA · SXM6 AC↗ 288 GBHBM3e Blackwell Ultra LLM inference, training and HPC ModalRunpod $6.94Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA GeForce RTX 4080 SUPER — RTX 4080 SUPER reference product view. Source: NVIDIA.GeForce RTX 4080 SUPERNVIDIA · PCIe add-in card↗ 16 GBGDDR6X Ada Lovelace AI inference, rendering and professional graphics Runpod $0.28Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA H100 NVL — Reference hardware; rental provider cooling configuration may differ.H100 NVLNVIDIA · PCIe / NVL↗ 94 GBHBM3 Hopper Memory-intensive inference RunpodVast.ai $2.59Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA H100 PCIe — Reference hardware; rental provider cooling configuration may differ.H100 PCIeNVIDIA · PCIe↗ 80 GBHBM2e Hopper Training and inference Runpod $1.99Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA H100 SXM — Reference hardware; rental provider cooling configuration may differ.H100 SXMNVIDIA · SXM5↗ 80 GBHBM3 Hopper Multi-GPU training and inference ModalRunpodVast.ai $2.3348Updated 2026-11-22 · 15:43 UTCOn-demand · Vast.aiPublished list price
NVIDIA H200 — NVIDIA H200 SXM5 single GPU module with heatsink and protective connector caps; 141 GB HBM3e reference.H200NVIDIA · SXM↗ 141 GBHBM3e Hopper Large-model inference and HPC ModalRunpodVast.ai $3.2909Updated 2026-11-22 · 15:43 UTCOn-demand · Vast.aiPublished list price
NVIDIA H200 NVL — H200 NVL reference product view. Source: NVIDIA.H200 NVLNVIDIA · PCIe NVL↗ 141 GBHBM3e Hopper LLM inference, training and HPC Runpod $3.79Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA L4 — Reference hardware; rental provider cooling configuration may differ.L4NVIDIA · PCIe↗ 24 GBGDDR6 Ada Lovelace Inference and video processing ModalRunpodVast.ai $0.1751Updated 2026-11-22 · 15:43 UTCOn-demand · Vast.aiPublished list price
NVIDIA L40 — Reference hardware; rental provider cooling configuration may differ.L40NVIDIA · PCIe↗ 48 GBGDDR6 Ada Lovelace Rendering and inference RunpodVast.ai $0.5081Updated 2026-11-22 · 15:43 UTCOn-demand · Vast.aiPublished list price
NVIDIA L40S — Reference hardware; rental provider cooling configuration may differ.L40SNVIDIA · PCIe↗ 48 GBGDDR6 Ada Lovelace Inference, fine-tuning and rendering CloudRiftModalRunpodVast.ai $0.5363Updated 2026-11-22 · 15:43 UTCOn-demand · Vast.aiPublished list price
NVIDIA RTX 2000 Ada — RTX 2000 Ada reference product view. Source: NVIDIA.RTX 2000 AdaNVIDIA · PCIe small-form-factor, dual slot↗ 16 GBGDDR6 Ada Lovelace AI inference, rendering and professional graphics Runpod $0.24Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX 3090 — PNY partner card; rental provider chassis and cooler may differ.RTX 3090NVIDIA · PCIe↗ 24 GBGDDR6X Ampere Development and inference RunpodSaladCloudVast.ai $0.09Updated 2026-11-22 · 15:43 UTCServerless · SaladCloudCPU/RAM included
NVIDIA RTX 4000 Ada — RTX 4000 Ada reference product view. Source: NVIDIA.RTX 4000 AdaNVIDIA · PCIe workstation, single slot↗ 20 GBGDDR6 Ada Lovelace AI inference, rendering and professional graphics Runpod $0.20Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX 4090 — PNY partner card; rental provider chassis and cooler may differ.RTX 4090NVIDIA · PCIe↗ 24 GBGDDR6X Ada Lovelace Image generation and inference CloudRiftRunpodSaladCloudVast.ai $0.16Updated 2026-11-22 · 15:43 UTCServerless · SaladCloudCPU/RAM included
NVIDIA RTX 5000 Ada — RTX 5000 Ada reference product view. Source: NVIDIA.RTX 5000 AdaNVIDIA · PCIe workstation, dual slot↗ 32 GBGDDR6 Ada Lovelace AI inference, rendering and professional graphics Runpod $0.49Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX 5090 — PNY partner card; rental provider chassis and cooler may differ.RTX 5090NVIDIA · PCIe↗ 32 GBGDDR7 Blackwell Image generation and inference CloudRiftRunpodSaladCloudVast.ai $0.25Updated 2026-11-22 · 15:43 UTCServerless · SaladCloudCPU/RAM included
NVIDIA RTX 6000 Ada — Reference hardware; rental provider cooling configuration may differ.RTX 6000 AdaNVIDIA · PCIe↗ 48 GBGDDR6 Ada Lovelace Professional rendering and inference Runpod $0.74Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX A2000 6GB — RTX A2000 6GB reference product view. Source: NVIDIA.RTX A2000 6GBNVIDIA · PCIe small-form-factor, dual slot↗ 6 GBGDDR6 Ampere AI inference, rendering and professional graphics Runpod $0.12Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX A4500 — RTX A4500 reference product view. Source: NVIDIA.RTX A4500NVIDIA · PCIe workstation, dual slot↗ 20 GBGDDR6 Ampere AI inference, rendering and professional graphics Runpod $0.19Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX A5000 — Reference hardware; rental provider cooling configuration may differ.RTX A5000NVIDIA · PCIe↗ 24 GBGDDR6 Ampere Development and rendering RunpodVast.ai $0.16Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX A6000 — Reference hardware; rental provider cooling configuration may differ.RTX A6000NVIDIA · PCIe↗ 48 GBGDDR6 Ampere Visualization and training RunpodVast.ai $0.33Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX PRO 4000 Blackwell — RTX PRO 4000 Blackwell reference product view. Source: NVIDIA.RTX PRO 4000 BlackwellNVIDIA · PCIe workstation, single slot↗ 24 GBGDDR7 Blackwell AI inference, rendering and professional graphics Runpod $0.57Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX PRO 4500 Blackwell — RTX PRO 4500 Blackwell reference product view. Source: NVIDIA.RTX PRO 4500 BlackwellNVIDIA · PCIe workstation, dual slot↗ 32 GBGDDR7 Blackwell AI inference, rendering and professional graphics Runpod $0.72Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX PRO 4500 Blackwell Server Edition — RTX PRO 4500 Blackwell Server Edition reference product view. Source: NVIDIA.RTX PRO 4500 Blackwell Server EditionNVIDIA · PCIe server, passive single slot↗ 32 GBGDDR7 Blackwell AI inference, rendering and professional graphics Runpod $0.72Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX PRO 5000 Blackwell 48GB — RTX PRO 5000 Blackwell 48GB reference product view. Source: NVIDIA.RTX PRO 5000 Blackwell 48GBNVIDIA · PCIe workstation, dual slot↗ 48 GBGDDR7 Blackwell AI inference, rendering and professional graphics Runpod $0.82Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX PRO 6000 Blackwell — Server Edition; rental provider configuration may differ.RTX PRO 6000 BlackwellNVIDIA · Provider variant; confirm edition↗ 96 GBGDDR7 Blackwell Rendering and large-model inference CloudRiftModalRunpod $1.69Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX PRO 6000 Blackwell Max-Q — RTX PRO 6000 Blackwell Max-Q reference product view. Source: NVIDIA.RTX PRO 6000 Blackwell Max-QNVIDIA · PCIe Max-Q Workstation↗ 96 GBGDDR7 Blackwell AI inference, rendering and professional graphics CloudRiftRunpod $1.64Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX PRO 6000 Blackwell Workstation — RTX PRO 6000 Blackwell Workstation reference product view. Source: NVIDIA.RTX PRO 6000 Blackwell WorkstationNVIDIA · PCIe Workstation↗ 96 GBGDDR7 Blackwell AI inference, rendering and professional graphics Runpod $1.69Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX PRO 6000 MIG 24GB — Parent GPU reference image for RTX PRO 6000 MIG 24GB; this MIG allocation is a partition, not a separate physical card.RTX PRO 6000 MIG 24GBNVIDIA · MIG partition — Server Edition 1g.24gb↗ 24 GBGDDR7 Blackwell AI inference, rendering and professional graphics Runpod $0.69USD / partition-hourUpdated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA RTX PRO 6000 MIG 48GB — Parent GPU reference image for RTX PRO 6000 MIG 48GB; this MIG allocation is a partition, not a separate physical card.RTX PRO 6000 MIG 48GBNVIDIA · MIG partition — Server Edition 2g.48gb↗ 48 GBGDDR7 Blackwell AI inference, rendering and professional graphics Runpod $1.39USD / partition-hourUpdated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
NVIDIA T4 — Reference hardware; rental provider cooling configuration may differ.T4NVIDIA · PCIe↗ 16 GBGDDR6 Turing Small-model inference and video Modal $0.5904Updated 2026-11-22 · 15:43 UTCServerless · ModalCPU/RAM: confirm with provider
NVIDIA V100 SXM2 16GB — V100 SXM2 16GB reference product view. Source: NVIDIA.V100 SXM2 16GBNVIDIA · SXM2↗ 16 GBHBM2 Volta LLM inference, training and HPC CloudRiftRunpod $0.23Updated 2026-11-22 · 15:43 UTCOn-demand · RunpodPublished list price
GTX 1070 — NVIDIA Founders Edition reference; partner cooling may differ.GTX 1070NVIDIA · PCIe add-in card↗ 8 GBGDDR5 Pascal Small CUDA workloads and legacy applications No offers yet Price unavailableView rental optionsPublished list price
GTX 1080 — Manufacturer reference image; exact provider board and cooling may differ.GTX 1080NVIDIA · PCIe add-in card↗ 8 GBGDDR5X Pascal Small CUDA workloads and legacy applications No offers yet Price unavailableView rental optionsPublished list price
GTX 1080 Ti — NVIDIA Founders Edition reference; partner cooling may differ.GTX 1080 TiNVIDIA · PCIe add-in card↗ 11 GBGDDR5X Pascal Small CUDA workloads and legacy applications No offers yet Price unavailableView rental optionsPublished list price
GTX 1650 — GeForce GTX 1650 Single Fan; partner/reference board appearance, not a rental-server guarantee.GTX 1650NVIDIA · PCIe add-in card↗ 4 GBGDDR5 Turing Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
GTX 1660 — GeForce GTX 1660 XLR8 Gaming Overclocked Edition Dual Fan; partner/reference board appearance, not a rental-server guarantee.GTX 1660NVIDIA · PCIe add-in card↗ 6 GBGDDR5 Turing Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
GTX 1660 Super — GeForce GTX 1660 SUPER™ Blower; partner/reference board appearance, not a rental-server guarantee.GTX 1660 SuperNVIDIA · PCIe add-in card↗ 6 GBGDDR6 Turing Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
GTX TITAN X — NVIDIA GTX TITAN X Maxwell reference card; three-quarter view.GTX TITAN XNVIDIA · PCIe add-in card↗ 12 GBGDDR5 Maxwell Small CUDA workloads and legacy applications No offers yet Price unavailableView rental optionsPublished list price
NVIDIA B300 MIG 34GB — Parent GPU reference image for B300 MIG 34GB; this MIG allocation is a partition, not a separate physical card.B300 MIG 34GBNVIDIA · MIG partition — B300 1g.34gb↗ 34 GBHBM3e Blackwell Ultra LLM inference, training and HPC No offers yet Price unavailableUSD / partition-hourView rental optionsPublished list price
NVIDIA RTX 4000 Ada SFF — RTX 4000 Ada SFF reference product view. Source: NVIDIA.RTX 4000 Ada SFFNVIDIA · PCIe small-form-factor, dual slot↗ 20 GBGDDR6 Ada Lovelace AI inference, rendering and professional graphics No offers yet Price unavailableView rental optionsPublished list price
NVIDIA Tesla V100 PCIe 16GB — Tesla V100 PCIe 16GB reference product view. Source: NVIDIA.Tesla V100 PCIe 16GBNVIDIA · PCIe↗ 16 GBHBM2 Volta LLM inference, training and HPC No offers yet Price unavailableView rental optionsPublished list price
NVIDIA V100 SXM3 32GB — NVIDIA V100 SXM family reference image, not an exact SXM3 board photograph. This record is the provider-listed SXM3 32 GB configuration.V100 SXM3 32GBNVIDIA · SXM3↗ 32 GBHBM2 Volta LLM inference, training and HPC CloudRift Price unavailableView rental optionsPublished list price
Quadro P2000 — NVIDIA Quadro P2000; partner/reference board appearance, not a rental-server guarantee.Quadro P2000NVIDIA · PCIe add-in card↗ 5 GBGDDR5 Pascal Small CUDA workloads and legacy applications No offers yet Price unavailableView rental optionsPublished list price
Quadro P4000 — NVIDIA Quadro P4000; partner/reference board appearance, not a rental-server guarantee.Quadro P4000NVIDIA · PCIe add-in card↗ 8 GBGDDR5 Pascal Small CUDA workloads and legacy applications No offers yet Price unavailableView rental optionsPublished list price
Quadro RTX 4000 — NVIDIA Quadro RTX 4000; partner/reference board appearance, not a rental-server guarantee.Quadro RTX 4000NVIDIA · PCIe add-in card↗ 8 GBGDDR6 Turing Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
Quadro RTX 5000 — NVIDIA Quadro RTX 5000; partner/reference board appearance, not a rental-server guarantee.Quadro RTX 5000NVIDIA · PCIe add-in card↗ 16 GBGDDR6 Turing Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 2060 — PNY GeForce RTX 2060 6GB Blower; partner/reference board appearance, not a rental-server guarantee.RTX 2060NVIDIA · PCIe add-in card↗ 6 GBGDDR6 Turing Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 2070 — Manufacturer reference image; exact provider board and cooling may differ.RTX 2070NVIDIA · PCIe add-in card↗ 8 GBGDDR6 Turing Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 2070 Super — PNY GeForce RTX 2070 Super 8GB Blower; partner/reference board appearance, not a rental-server guarantee.RTX 2070 SuperNVIDIA · PCIe add-in card↗ 8 GBGDDR6 Turing Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 2080 Ti — GeForce RTX 2080 Ti 11GB Blower; partner/reference board appearance, not a rental-server guarantee.RTX 2080 TiNVIDIA · PCIe add-in card↗ 11 GBGDDR6 Turing Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 3060 — PNY GeForce RTX 3060 12GB XLR8 Gaming REVEL EPIC-X RGB Dual Fan; partner/reference board appearance, not a rental-server guarantee.RTX 3060NVIDIA · PCIe add-in card↗ 12 GBGDDR6 Ampere Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 3060 Laptop — NVIDIA GN20-E3-A1 (GA106) GPU chip used by GeForce RTX 3060 Laptop; BGA package reference.RTX 3060 LaptopNVIDIA · Laptop GPU (not a standalone PCIe card)↗ 6 GBGDDR6 Ampere Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 3060 Ti — PNY GeForce RTX 3060 Ti 8GB XLR8 Gaming REVEL EPIC-X RGB Dual Fan Edition; partner/reference board appearance, not a rental-server guarantee.RTX 3060 TiNVIDIA · PCIe add-in card↗ 8 GBGDDR6 Ampere Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 3070 — XLR8 Gaming GeForce RTX 3070 UPRISING Dual Fan Edition; partner/reference board appearance, not a rental-server guarantee.RTX 3070NVIDIA · PCIe add-in card↗ 8 GBGDDR6 Ampere Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 3080 — XLR8 Gaming GeForce RTX 3080 REVEL EPIC-X RGB Triple Fan Edition; partner/reference board appearance, not a rental-server guarantee.RTX 3080NVIDIA · PCIe add-in card↗ 12 GBGDDR6X Ampere Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 3080 — XLR8 Gaming GeForce RTX 3080 REVEL EPIC-X RGB Triple Fan Edition; partner/reference board appearance, not a rental-server guarantee.RTX 3080NVIDIA · PCIe add-in card↗ 12 GBGDDR6X Ampere Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 3080 Ti — XLR8 Gaming GeForce RTX 3080 Ti REVEL EPIC-X RGB Triple Fan Edition; partner/reference board appearance, not a rental-server guarantee.RTX 3080 TiNVIDIA · PCIe add-in card↗ 12 GBGDDR6X Ampere Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 3090 Ti — PNY GeForce RTX 3090 Ti 24GB XLR8 Gaming UPRISING EPIC-X RGB Overclocked Triple Fan; partner/reference board appearance, not a rental-server guarantee.RTX 3090 TiNVIDIA · PCIe add-in card↗ 24 GBGDDR6X Ampere Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 4060 — PNY GeForce RTX™ 4060 8GB XLR8 Gaming VERTO RGB Triple Fan Graphics Card DLSS 3 (OC & Non-OC); partner/reference board appearance, not a rental-server guarantee.RTX 4060NVIDIA · PCIe add-in card↗ 8 GBGDDR6 Ada Lovelace Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 4060 Ti — PNY GeForce RTX™ 4060 Ti XLR8 Gaming VERTO RGB Triple Fan Graphics Card DLSS 3 (8GB & 16GB) (OC & Non-OC); partner/reference board appearance, not a rental-server guarantee.RTX 4060 TiNVIDIA · PCIe add-in card↗ 16 GBGDDR6 Ada Lovelace Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 4060 Ti — PNY GeForce RTX™ 4060 Ti XLR8 Gaming VERTO RGB Triple Fan Graphics Card DLSS 3 (8GB & 16GB) (OC & Non-OC); partner/reference board appearance, not a rental-server guarantee.RTX 4060 TiNVIDIA · PCIe add-in card↗ 16 GBGDDR6 Ada Lovelace Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 4070 — PNY GeForce RTX™ 4070 12GB XLR8 Gaming VERTO EPIC-X RGB; partner/reference board appearance, not a rental-server guarantee.RTX 4070NVIDIA · PCIe add-in card↗ 12 GBGDDR6 Ada Lovelace Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 4070 Ti — PNY GeForce RTX™ 4070 Ti 12GB XLR8 Gaming VERTO EPIC-X RGB Triple Fan; partner/reference board appearance, not a rental-server guarantee.RTX 4070 TiNVIDIA · PCIe add-in card↗ 12 GBGDDR6X Ada Lovelace Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 4070 Ti Super — PNY GeForce RTX™ 4070 Ti SUPER™ 16GB VERTO Overclocked Triple Fan; partner/reference board appearance, not a rental-server guarantee.RTX 4070 Ti SuperNVIDIA · PCIe add-in card↗ 16 GBGDDR6X Ada Lovelace Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 4080 — PNY GeForce RTX™ 4080 16GB XLR8 Gaming VERTO EPIC-X RGB Triple Fan; partner/reference board appearance, not a rental-server guarantee.RTX 4080NVIDIA · PCIe add-in card↗ 16 GBGDDR6X Ada Lovelace Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 5060 — PNY GeForce RTX™ 5060 8GB Dual Fan; partner/reference board appearance, not a rental-server guarantee.RTX 5060NVIDIA · PCIe add-in card↗ 8 GBGDDR7 Blackwell Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 5060 Ti — PNY GeForce RTX™ 5060 Ti 16GB Dual Fan; partner/reference board appearance, not a rental-server guarantee.RTX 5060 TiNVIDIA · PCIe add-in card↗ 16 GBGDDR7 Blackwell Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 5070 — PNY GeForce RTX™ 5070 12GB Triple Fan; partner/reference board appearance, not a rental-server guarantee.RTX 5070NVIDIA · PCIe add-in card↗ 12 GBGDDR7 Blackwell Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 5070 Ti — PNY GeForce RTX™ 5070 Ti 16GB Triple Fan Plus; partner/reference board appearance, not a rental-server guarantee.RTX 5070 TiNVIDIA · PCIe add-in card↗ 16 GBGDDR7 Blackwell Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 5080 — PNY GeForce RTX™ 5080 16GB Triple Fan; partner/reference board appearance, not a rental-server guarantee.RTX 5080NVIDIA · PCIe add-in card↗ 16 GBGDDR7 Blackwell Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX 5090 Laptop — NVIDIA GB203 notebook GPU package reference used by RTX 5090 Laptop; chip image, not a desktop graphics card.RTX 5090 LaptopNVIDIA · Laptop GPU (not a standalone PCIe card)↗ 24 GBGDDR7 Blackwell Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
RTX A4000 — NVIDIA RTX A4000; partner/reference board appearance, not a rental-server guarantee.RTX A4000NVIDIA · PCIe add-in card↗ 16 GBGDDR6 Ampere Rendering, graphics and memory-budgeted inference No offers yet Price unavailableView rental optionsPublished list price
Tesla P100 — Manufacturer reference image; exact provider board and cooling may differ.Tesla P100NVIDIA · PCIe add-in card↗ 16 GBHBM2 Pascal Small CUDA workloads and legacy applications No offers yet Price unavailableView rental optionsPublished list price
Tesla P40 — Manufacturer reference image; exact provider board and cooling may differ.Tesla P40NVIDIA · PCIe add-in card↗ 24 GBGDDR5 Pascal Small CUDA workloads and legacy applications No offers yet Price unavailableView rental optionsPublished list price
Titan Xp — Manufacturer reference image; exact provider board and cooling may differ.Titan XpNVIDIA · PCIe add-in card↗ 12 GBGDDR5X Pascal Small CUDA workloads and legacy applications No offers yet Price unavailableView rental optionsPublished list price
How to compare these prices +

“No reservation” includes on-demand and serverless offers. The rental type and provider are shown next to each price. Serverless CPU/RAM inclusion varies by provider; check the note beside the price. Storage may be extra. Reserved prices require a commitment; spot capacity can be interrupted. Prices are per GPU-hour; MIG rows are per partition-hour. Prices follow your filters. A multi-GPU minimum is shown when required. Quote-only and unavailable offers are never used as starting prices. Sources and check dates are available on each GPU page.