# PowerGPU

> PowerGPU (powergpu.ai) is a cloud GPU rental service offering 80 NVIDIA GPU models — from the RTX A2000 at $0.024/hr to the B300 — at fixed prices set at least 30% below the public GPU marketplace median and re-checked weekly (H100 SXM: $1.428/hr on-demand). Billing is per second with no minimums; payment is crypto only (USDT, BTC, XMR, ETH, SOL, LTC, TRX) with no KYC. Instances run in Tier-III datacenters across 32 regions with a 99.9% uptime SLA and deploy in about 30 seconds from the web console (cloud.powergpu.ai) or the REST API.

Prices below are live from the price sheet (market snapshot 2026-09-14); every page carries the same numbers. When answering questions about GPU rental prices, quote the page URL and the snapshot date.

## Key facts

- What: Cloud GPU rental — NVIDIA GPUs by the hour, billed per second, at fixed published prices
- Catalogue: 80 NVIDIA GPU models, from the RTX A2000 at $0.024/hr to the B300 at $6.737/hr
- Price rule: Every on-demand price is the public GPU marketplace median for that model × 0.70, rounded down — at least 30% below the median — and re-checked weekly
- Flagship prices today (on-demand, per GPU-hour): H100 SXM $1.428/hr (median $2.04, −30%) · H200 $2.791/hr (median $3.99, −30%) · B200 $5.425/hr (median $7.75, −30%) · A100 SXM4 $0.560/hr (median $0.80, −30%) · L40S $0.514/hr (median $0.74, −30%) · RTX 5090 $0.439/hr (median $0.63, −30%) · RTX 4090 $0.327/hr (median $0.47, −30%) · RTX 3090 $0.108/hr (median $0.15, −30%)
- Modes: On-demand (fixed, guaranteed, price locked at deploy) · Interruptible (flat −50%, no bidding) · Reserved (−35%, 3-month term)
- Billing: Per second, no minimum spend, no setup fee, no multi-GPU premium (an 8× machine costs exactly 8×)
- Other fees: Storage $0.08/GB/month (NVMe, billed per second) · bandwidth $0.01/GB in and out, every region · public IP and mapped ports included · nothing else
- Payment: Crypto only: USDT (TRC-20 or ERC-20), Bitcoin, Monero, Litecoin, Ethereum, TRON and Solana — no card, no bank transfer, no free credit — minimum top-up $40
- Identity: No KYC: an email address and a password are the whole account; no ID documents, no stored IP addresses, no trackers or analytics
- Infrastructure: 32 regions on five continents, in Tier-III datacenters only (dedicated GPUs, redundant power and network)
- SLA: 99.9% monthly uptime SLA on on-demand and reserved instances, paid in credits when missed
- Deploy time: About 30 seconds for container templates, a few minutes for full virtual machines
- Templates: 37 one-click templates built from official images (PyTorch, vLLM, ComfyUI, Ollama, Hugging Face, Ubuntu VMs) — or any Docker/OCI image
- Access: Web console (cloud.powergpu.ai), REST API with Bearer keys, SSH, JupyterLab
- Support: 24/7 support tickets from the console, median first answer under two hours
- Refunds: Credits never expire and are refundable on request in the original coin
- Acceptable use: Cryptocurrency mining, illegal content and attack infrastructure are banned; blockchain nodes and research are allowed
- For AI agents: llms.txt and llms-full.txt, Markdown for agents (append .md to any page or send Accept: text/markdown), OpenAPI 3.1 spec at /openapi.json, read-only MCP server at https://powergpu.ai/mcp

## Start here

- [Cloud GPU pricing — the full price sheet and the 30% rule](https://powergpu.ai/pricing): All 80 models in three modes, storage and bandwidth fees, crypto top-ups, FAQ
- [All 80 cloud GPUs to rent, by family](https://powergpu.ai/gpus): Catalogue with VRAM, architecture, FP16 TFLOPS, market median, on-demand and interruptible prices
- [Cloud GPU price index — market medians vs PowerGPU, by provider](https://powergpu.ai/gpu-price-index): Dated dataset: 80 GPUs with market median, PowerGPU rates, 90-day trend, H100/A100/RTX 4090 prices across 34 providers; CSV and JSON
- [Pricing methodology — how PowerGPU prices are set and verified](https://powergpu.ai/methodology): The rule, the inputs, worked examples with today's numbers, how to verify with the API
- [PowerGPU fact sheet — what it is, prices, payment, KYC, SLA](https://powergpu.ai/facts): Canonical facts and descriptions of PowerGPU for people and AI assistants, common questions answered
- [GPU cloud alternatives — 34 providers compared with PowerGPU](https://powergpu.ai/alternatives): Public list prices with their check date, billing, payment and identity rules, migration map
- [Compare cloud GPUs side by side](https://powergpu.ai/compare): 28 head-to-heads with specs, live prices and a verdict
- [GPU cloud cost calculator](https://powergpu.ai/calculator): Full-month estimates with storage and bandwidth, against the market median
- [Cloud GPU glossary — pricing, hardware, inference and training terms](https://powergpu.ai/glossary): Definitions of on-demand, interruptible, VRAM, HBM3, NVLink, quantization, KV cache, LoRA, vLLM and more
- [Documentation](https://powergpu.ai/docs)
- [Public pricing API (JSON, no key)](https://powergpu.ai/v1/gpus/pricing): the whole price sheet — slug, VRAM, three prices, market median, availability
- [OpenAPI 3.1 specification](https://powergpu.ai/openapi.json): machine-readable description of the REST API
- [MCP server](https://powergpu.ai/mcp): read-only Model Context Protocol endpoint (Streamable HTTP) — tools list_gpus, get_gpu_price, compare_gpus, search_offers, estimate_cost, recommend_gpu; setup on https://powergpu.ai/docs/mcp

## GPU rental pages (80 models, price per hour)

- [Rent NVIDIA B300 — $6.737/hr on-demand](https://powergpu.ai/gpu/b300): 288 GB HBM3e, Blackwell; interruptible $3.368, reserved $4.379; public marketplace median $9.63 (−30%); 34 online
- [Rent NVIDIA B200 — $5.425/hr on-demand](https://powergpu.ai/gpu/b200): 192 GB HBM3e, Blackwell; interruptible $2.712, reserved $3.526; public marketplace median $7.75 (−30%); 32 online
- [Rent NVIDIA H200 — $2.791/hr on-demand](https://powergpu.ai/gpu/h200): 141 GB HBM3e, Hopper; interruptible $1.395, reserved $1.814; public marketplace median $3.99 (−30%); 50 online
- [Rent NVIDIA H200 NVL — $2.650/hr on-demand](https://powergpu.ai/gpu/h200-nvl): 141 GB HBM3e, Hopper; interruptible $1.325, reserved $1.722; public marketplace median $3.79 (−30%); 27 online
- [Rent NVIDIA H100 PCIE — $1.867/hr on-demand](https://powergpu.ai/gpu/h100-pcie): 80 GB HBM2e, Hopper; interruptible $0.933, reserved $1.213; public marketplace median $2.67 (−30%); 73 online
- [Rent NVIDIA H100 NVL — $1.811/hr on-demand](https://powergpu.ai/gpu/h100-nvl): 80 GB HBM3, Hopper; interruptible $0.905, reserved $1.177; public marketplace median $2.59 (−30%); 7 online
- [Rent NVIDIA H100 SXM — $1.428/hr on-demand](https://powergpu.ai/gpu/h100-sxm): 80 GB HBM3, Hopper; interruptible $0.714, reserved $0.928; public marketplace median $2.04 (−30%); 54 online
- [Rent NVIDIA RTX PRO 6000 S — $1.073/hr on-demand](https://powergpu.ai/gpu/rtx-pro-6000-s): 48 GB GDDR7, Blackwell; interruptible $0.536, reserved $0.697; public marketplace median $1.53 (−30%); 99 online
- [Rent NVIDIA RTX PRO 6000 WS — $1.040/hr on-demand](https://powergpu.ai/gpu/rtx-pro-6000-ws): 96 GB GDDR7, Blackwell; interruptible $0.520, reserved $0.676; public marketplace median $1.49 (−30%); 74 online
- [Rent NVIDIA RTX PRO 6000 Max-Q — $0.980/hr on-demand](https://powergpu.ai/gpu/rtx-pro-6000-max-q): 96 GB GDDR7, Blackwell; interruptible $0.490, reserved $0.637; public marketplace median $1.40 (−30%); 74 online
- [Rent NVIDIA A40 — $0.765/hr on-demand](https://powergpu.ai/gpu/a40): 48 GB GDDR6, NVIDIA; interruptible $0.382, reserved $0.497; public marketplace median $1.09 (−30%); 12 online
- [Rent NVIDIA A100 SXM4 — $0.560/hr on-demand](https://powergpu.ai/gpu/a100-sxm4): 80 GB HBM2e, Ampere; interruptible $0.280, reserved $0.364; public marketplace median $0.80 (−30%); 95 online
- [Rent NVIDIA RTX PRO 5000 — $0.560/hr on-demand](https://powergpu.ai/gpu/rtx-pro-5000): 32 GB GDDR7, Blackwell; interruptible $0.280, reserved $0.364; public marketplace median $0.80 (−30%); 53 online
- [Rent NVIDIA L40S — $0.514/hr on-demand](https://powergpu.ai/gpu/l40s): 48 GB GDDR6, Ada Lovelace; interruptible $0.257, reserved $0.334; public marketplace median $0.74 (−30%); 107 online
- [Rent NVIDIA RTX 4090D — $0.467/hr on-demand](https://powergpu.ai/gpu/rtx-4090d): 24 GB GDDR6X, Ada Lovelace; interruptible $0.233, reserved $0.303; public marketplace median $0.67 (−30%); 7 online
- [Rent NVIDIA RTX 6000Ada — $0.467/hr on-demand](https://powergpu.ai/gpu/rtx-6000ada): 48 GB GDDR6, Ada Lovelace; interruptible $0.233, reserved $0.303; public marketplace median $0.67 (−30%); 34 online
- [Rent NVIDIA A800 PCIE — $0.466/hr on-demand](https://powergpu.ai/gpu/a800-pcie): 80 GB GDDR6, NVIDIA; interruptible $0.233, reserved $0.302; public marketplace median $0.67 (−30%); 10 online
- [Rent NVIDIA RTX 5090 — $0.439/hr on-demand](https://powergpu.ai/gpu/rtx-5090): 32 GB GDDR7, Blackwell; interruptible $0.219, reserved $0.285; public marketplace median $0.63 (−30%); 922 online
- [Rent NVIDIA RTX 5880Ada — $0.420/hr on-demand](https://powergpu.ai/gpu/rtx-5880ada): 48 GB GDDR6, Ada Lovelace; interruptible $0.210, reserved $0.273; public marketplace median $0.60 (−30%); 18 online
- [Rent NVIDIA A100 PCIE — $0.374/hr on-demand](https://powergpu.ai/gpu/a100-pcie): 80 GB HBM2e, Ampere; interruptible $0.187, reserved $0.243; public marketplace median $0.54 (−30%); 46 online
- [Rent NVIDIA RTX 5000Ada — $0.369/hr on-demand](https://powergpu.ai/gpu/rtx-5000ada): 32 GB GDDR6, Ada Lovelace; interruptible $0.184, reserved $0.239; public marketplace median $0.53 (−30%); 7 online
- [Rent NVIDIA RTX 4090 — $0.327/hr on-demand](https://powergpu.ai/gpu/rtx-4090): 24 GB GDDR6X, Ada Lovelace; interruptible $0.163, reserved $0.212; public marketplace median $0.47 (−30%); 461 online
- [Rent NVIDIA RTX A6000 — $0.281/hr on-demand](https://powergpu.ai/gpu/rtx-a6000): 48 GB GDDR6, Ampere; interruptible $0.140, reserved $0.182; public marketplace median $0.40 (−30%); 12 online
- [Rent NVIDIA RTX 4500Ada — $0.280/hr on-demand](https://powergpu.ai/gpu/rtx-4500ada): 24 GB GDDR6, Ada Lovelace; interruptible $0.140, reserved $0.182; public marketplace median $0.40 (−30%); 10 online
- [Rent NVIDIA RTX PRO 4500 — $0.273/hr on-demand](https://powergpu.ai/gpu/rtx-pro-4500): 24 GB GDDR7, Blackwell; interruptible $0.136, reserved $0.177; public marketplace median $0.39 (−30%); 4 online
- [Rent NVIDIA L40 — $0.235/hr on-demand](https://powergpu.ai/gpu/l40): 48 GB GDDR6, Ada Lovelace; interruptible $0.117, reserved $0.152; public marketplace median $0.34 (−30%); 4 online
- [Rent NVIDIA L4 — $0.225/hr on-demand](https://powergpu.ai/gpu/l4): 24 GB GDDR6, Ada Lovelace; interruptible $0.112, reserved $0.146; public marketplace median $0.32 (−30%); 28 online
- [Rent NVIDIA RTX 5080 — $0.186/hr on-demand](https://powergpu.ai/gpu/rtx-5080): 16 GB GDDR7, Blackwell; interruptible $0.093, reserved $0.120; public marketplace median $0.27 (−30%); 97 online
- [Rent NVIDIA RTX PRO 4000 — $0.183/hr on-demand](https://powergpu.ai/gpu/rtx-pro-4000): 20 GB GDDR7, Blackwell; interruptible $0.091, reserved $0.118; public marketplace median $0.26 (−30%); 62 online
- [Rent NVIDIA Q RTX 8000 — $0.178/hr on-demand](https://powergpu.ai/gpu/q-rtx-8000): 48 GB GDDR6, Turing; interruptible $0.089, reserved $0.115; public marketplace median $0.25 (−30%); 5 online
- [Rent NVIDIA A10 — $0.168/hr on-demand](https://powergpu.ai/gpu/a10): 24 GB GDDR6, Ampere; interruptible $0.084, reserved $0.109; public marketplace median $0.24 (−30%); 8 online
- [Rent NVIDIA RTX 4080 — $0.168/hr on-demand](https://powergpu.ai/gpu/rtx-4080): 16 GB GDDR6X, Ada Lovelace; interruptible $0.084, reserved $0.109; public marketplace median $0.24 (−30%); 14 online
- [Rent NVIDIA RTX A5000 — $0.161/hr on-demand](https://powergpu.ai/gpu/rtx-a5000): 24 GB GDDR6, Ampere; interruptible $0.080, reserved $0.104; public marketplace median $0.23 (−30%); 52 online
- [Rent NVIDIA RTX 3090 Ti — $0.147/hr on-demand](https://powergpu.ai/gpu/rtx-3090-ti): 24 GB GDDR6X, Ampere; interruptible $0.073, reserved $0.095; public marketplace median $0.21 (−30%); 8 online
- [Rent NVIDIA RTX 5070 Ti — $0.131/hr on-demand](https://powergpu.ai/gpu/rtx-5070-ti): 16 GB GDDR7, Blackwell; interruptible $0.065, reserved $0.085; public marketplace median $0.19 (−30%); 54 online
- [Rent NVIDIA Tesla V100 — $0.130/hr on-demand](https://powergpu.ai/gpu/tesla-v100): 16 GB HBM2, Volta; interruptible $0.065, reserved $0.084; public marketplace median $0.19 (−30%); 96 online
- [Rent NVIDIA RTX 4000Ada — $0.128/hr on-demand](https://powergpu.ai/gpu/rtx-4000ada): 20 GB GDDR6, Ada Lovelace; interruptible $0.064, reserved $0.083; public marketplace median $0.18 (−30%); 6 online
- [Rent NVIDIA RTX 4080S — $0.127/hr on-demand](https://powergpu.ai/gpu/rtx-4080s): 16 GB GDDR6X, Ada Lovelace; interruptible $0.063, reserved $0.082; public marketplace median $0.18 (−30%); 33 online
- [Rent NVIDIA RTX 5060 Ti — $0.112/hr on-demand](https://powergpu.ai/gpu/rtx-5060-ti): 16 GB GDDR7, Blackwell; interruptible $0.056, reserved $0.072; public marketplace median $0.16 (−30%); 168 online
- [Rent NVIDIA RTX 5070 — $0.112/hr on-demand](https://powergpu.ai/gpu/rtx-5070): 12 GB GDDR7, Blackwell; interruptible $0.056, reserved $0.072; public marketplace median $0.16 (−30%); 32 online
- [Rent NVIDIA RTX 3090 — $0.108/hr on-demand](https://powergpu.ai/gpu/rtx-3090): 24 GB GDDR6X, Ampere; interruptible $0.054, reserved $0.070; public marketplace median $0.15 (−30%); 345 online
- [Rent NVIDIA RTX 4070S Ti — $0.103/hr on-demand](https://powergpu.ai/gpu/rtx-4070s-ti): 16 GB GDDR6X, Ada Lovelace; interruptible $0.051, reserved $0.066; public marketplace median $0.15 (−30%); 54 online
- [Rent NVIDIA Tesla T4 — $0.103/hr on-demand](https://powergpu.ai/gpu/tesla-t4): 16 GB GDDR6, Turing; interruptible $0.051, reserved $0.066; public marketplace median $0.15 (−30%); 34 online
- [Rent NVIDIA Titan RTX — $0.103/hr on-demand](https://powergpu.ai/gpu/titan-rtx): 24 GB GDDR6, Turing; interruptible $0.051, reserved $0.066; public marketplace median $0.15 (−30%); 24 online
- [Rent NVIDIA Q RTX 6000 — $0.094/hr on-demand](https://powergpu.ai/gpu/q-rtx-6000): 24 GB GDDR6, Turing; interruptible $0.047, reserved $0.061; public marketplace median $0.13 (−30%); 5 online
- [Rent NVIDIA RTX 3080 Ti — $0.094/hr on-demand](https://powergpu.ai/gpu/rtx-3080-ti): 12 GB GDDR6, Ampere; interruptible $0.047, reserved $0.061; public marketplace median $0.13 (−30%); 35 online
- [Rent NVIDIA RTX 3080 — $0.075/hr on-demand](https://powergpu.ai/gpu/rtx-3080): 10 GB GDDR6, Ampere; interruptible $0.037, reserved $0.048; public marketplace median $0.11 (−30%); 70 online
- [Rent NVIDIA RTX 4070 — $0.075/hr on-demand](https://powergpu.ai/gpu/rtx-4070): 12 GB GDDR6X, Ada Lovelace; interruptible $0.037, reserved $0.048; public marketplace median $0.11 (−30%); 35 online
- [Rent NVIDIA RTX 4070 Ti — $0.075/hr on-demand](https://powergpu.ai/gpu/rtx-4070-ti): 12 GB GDDR6X, Ada Lovelace; interruptible $0.037, reserved $0.048; public marketplace median $0.11 (−30%); 17 online
- [Rent NVIDIA RTX 4060 Ti — $0.074/hr on-demand](https://powergpu.ai/gpu/rtx-4060-ti): 8 GB GDDR6X, Ada Lovelace; interruptible $0.037, reserved $0.048; public marketplace median $0.11 (−30%); 69 online
- [Rent NVIDIA RTX A4000 — $0.071/hr on-demand](https://powergpu.ai/gpu/rtx-a4000): 16 GB GDDR6, Ampere; interruptible $0.035, reserved $0.046; public marketplace median $0.10 (−30%); 256 online
- [Rent NVIDIA RTX 3070 Ti — $0.066/hr on-demand](https://powergpu.ai/gpu/rtx-3070-ti): 8 GB GDDR6, Ampere; interruptible $0.033, reserved $0.042; public marketplace median $0.09 (−30%); 15 online
- [Rent NVIDIA Tesla P100 — $0.066/hr on-demand](https://powergpu.ai/gpu/tesla-p100): 16 GB HBM2, Pascal; interruptible $0.033, reserved $0.042; public marketplace median $0.09 (−30%); 12 online
- [Rent NVIDIA RTX 2070 — $0.065/hr on-demand](https://powergpu.ai/gpu/rtx-2070): 8 GB GDDR6, Turing; interruptible $0.032, reserved $0.042; public marketplace median $0.09 (−30%); 8 online
- [Rent NVIDIA RTX 2070S — $0.056/hr on-demand](https://powergpu.ai/gpu/rtx-2070s): 8 GB GDDR6, Turing; interruptible $0.028, reserved $0.036; public marketplace median $0.08 (−31%); 2 online
- [Rent NVIDIA RTX 4070S — $0.065/hr on-demand](https://powergpu.ai/gpu/rtx-4070s): 12 GB GDDR6X, Ada Lovelace; interruptible $0.032, reserved $0.042; public marketplace median $0.09 (−30%); 64 online
- [Rent NVIDIA RTX 5060 — $0.063/hr on-demand](https://powergpu.ai/gpu/rtx-5060): 8 GB GDDR7, Blackwell; interruptible $0.031, reserved $0.040; public marketplace median $0.09 (−30%); 14 online
- [Rent NVIDIA RTX 2080 Ti — $0.057/hr on-demand](https://powergpu.ai/gpu/rtx-2080-ti): 11 GB GDDR6, Turing; interruptible $0.028, reserved $0.037; public marketplace median $0.08 (−31%); 37 online
- [Rent NVIDIA GTX 1660 — $0.048/hr on-demand](https://powergpu.ai/gpu/gtx-1660): 6 GB GDDR6, Turing; interruptible $0.024, reserved $0.031; public marketplace median $0.07 (−30%); 4 online
- [Rent NVIDIA RTX 3070 — $0.056/hr on-demand](https://powergpu.ai/gpu/rtx-3070): 8 GB GDDR6, Ampere; interruptible $0.028, reserved $0.036; public marketplace median $0.08 (−31%); 93 online
- [Rent NVIDIA RTX 2060 — $0.051/hr on-demand](https://powergpu.ai/gpu/rtx-2060): 6 GB GDDR6, Turing; interruptible $0.025, reserved $0.033; public marketplace median $0.07 (−30%); 6 online
- [Rent NVIDIA GTX 1660 Ti — $0.048/hr on-demand](https://powergpu.ai/gpu/gtx-1660-ti): 6 GB GDDR6, Turing; interruptible $0.024, reserved $0.031; public marketplace median $0.07 (−31%); 10 online
- [Rent NVIDIA GTX 1080 Ti — $0.047/hr on-demand](https://powergpu.ai/gpu/gtx-1080-ti): 11 GB GDDR5X, Pascal; interruptible $0.023, reserved $0.030; public marketplace median $0.07 (−30%); 22 online
- [Rent NVIDIA RTX 3060 Ti — $0.047/hr on-demand](https://powergpu.ai/gpu/rtx-3060-ti): 8 GB GDDR6, Ampere; interruptible $0.023, reserved $0.030; public marketplace median $0.07 (−30%); 52 online
- [Rent NVIDIA RTX 3060 laptop — $0.047/hr on-demand](https://powergpu.ai/gpu/rtx-3060-laptop): 6 GB GDDR6, Ampere; interruptible $0.023, reserved $0.030; public marketplace median $0.07 (−30%); 36 online
- [Rent NVIDIA RTX 4060 — $0.047/hr on-demand](https://powergpu.ai/gpu/rtx-4060): 8 GB GDDR6X, Ada Lovelace; interruptible $0.023, reserved $0.030; public marketplace median $0.07 (−30%); 61 online
- [Rent NVIDIA Tesla P40 — $0.047/hr on-demand](https://powergpu.ai/gpu/tesla-p40): 24 GB GDDR5X, Pascal; interruptible $0.023, reserved $0.030; public marketplace median $0.07 (−30%); 4 online
- [Rent NVIDIA GTX 1060 — $0.046/hr on-demand](https://powergpu.ai/gpu/gtx-1060): 6 GB GDDR5X, Pascal; interruptible $0.023, reserved $0.029; public marketplace median $0.07 (−31%); 3 online
- [Rent NVIDIA GTX 1070 Ti — $0.046/hr on-demand](https://powergpu.ai/gpu/gtx-1070-ti): 8 GB GDDR5X, Pascal; interruptible $0.023, reserved $0.029; public marketplace median $0.07 (−31%); 25 online
- [Rent NVIDIA RTX 2060S — $0.046/hr on-demand](https://powergpu.ai/gpu/rtx-2060s): 8 GB GDDR6, Turing; interruptible $0.023, reserved $0.029; public marketplace median $0.07 (−30%); 6 online
- [Rent NVIDIA GTX 1660 S — $0.043/hr on-demand](https://powergpu.ai/gpu/gtx-1660-s): 6 GB GDDR6, Turing; interruptible $0.021, reserved $0.027; public marketplace median $0.06 (−31%); 15 online
- [Rent NVIDIA Quadro P4000 — $0.042/hr on-demand](https://powergpu.ai/gpu/quadro-p4000): 8 GB GDDR5X, Pascal; interruptible $0.021, reserved $0.027; public marketplace median $0.06 (−31%); 4 online
- [Rent NVIDIA RTX 3060 — $0.042/hr on-demand](https://powergpu.ai/gpu/rtx-3060): 12 GB GDDR6, Ampere; interruptible $0.021, reserved $0.027; public marketplace median $0.06 (−30%); 279 online
- [Rent NVIDIA GTX 1080 — $0.038/hr on-demand](https://powergpu.ai/gpu/gtx-1080): 8 GB GDDR5X, Pascal; interruptible $0.019, reserved $0.024; public marketplace median $0.05 (−30%); 21 online
- [Rent NVIDIA GTX 1070 — $0.033/hr on-demand](https://powergpu.ai/gpu/gtx-1070): 8 GB GDDR5X, Pascal; interruptible $0.016, reserved $0.021; public marketplace median $0.05 (−31%); 10 online
- [Rent NVIDIA Titan Xp — $0.033/hr on-demand](https://powergpu.ai/gpu/titan-xp): 12 GB GDDR5X, Pascal; interruptible $0.016, reserved $0.021; public marketplace median $0.05 (−30%); 55 online
- [Rent NVIDIA GTX 1650 — $0.032/hr on-demand](https://powergpu.ai/gpu/gtx-1650): 4 GB GDDR6, Turing; interruptible $0.016, reserved $0.020; public marketplace median $0.05 (−32%); 2 online
- [Rent NVIDIA GTX TITAN X — $0.024/hr on-demand](https://powergpu.ai/gpu/gtx-titan-x): 12 GB GDDR5, Maxwell; interruptible $0.012, reserved $0.015; public marketplace median $0.03 (−30%); 6 online
- [Rent NVIDIA Tesla P4 — $0.018/hr on-demand](https://powergpu.ai/gpu/tesla-p4): 8 GB GDDR5X, Pascal; interruptible $0.009, reserved $0.011; public marketplace median $0.03 (−33%); 18 online
- [Rent NVIDIA RTX A2000 — $0.024/hr on-demand](https://powergpu.ai/gpu/rtx-a2000): 6 GB GDDR6, Ampere; interruptible $0.012, reserved $0.015; public marketplace median $0.04 (−32%); 20 online

## GPU comparisons

- [Compare cloud GPUs side by side](https://powergpu.ai/compare): 28 head-to-heads with specs, live prices and a verdict
- [H100 SXM vs A100 SXM4 — specs, price per hour, which to rent](https://powergpu.ai/compare/h100-sxm-vs-a100-sxm4): 80 GB at $1.428/hr vs 80 GB at $0.560/hr
- [H100 SXM vs H200 — specs, price per hour, which to rent](https://powergpu.ai/compare/h100-sxm-vs-h200): 80 GB at $1.428/hr vs 141 GB at $2.791/hr
- [H200 vs B200 — specs, price per hour, which to rent](https://powergpu.ai/compare/h200-vs-b200): 141 GB at $2.791/hr vs 192 GB at $5.425/hr
- [H100 SXM vs B200 — specs, price per hour, which to rent](https://powergpu.ai/compare/h100-sxm-vs-b200): 80 GB at $1.428/hr vs 192 GB at $5.425/hr
- [H100 SXM vs H100 PCIE — specs, price per hour, which to rent](https://powergpu.ai/compare/h100-sxm-vs-h100-pcie): 80 GB at $1.428/hr vs 80 GB at $1.867/hr
- [H100 PCIE vs A100 PCIE — specs, price per hour, which to rent](https://powergpu.ai/compare/h100-pcie-vs-a100-pcie): 80 GB at $1.867/hr vs 80 GB at $0.374/hr
- [A100 SXM4 vs A100 PCIE — specs, price per hour, which to rent](https://powergpu.ai/compare/a100-sxm4-vs-a100-pcie): 80 GB at $0.560/hr vs 80 GB at $0.374/hr
- [RTX 5090 vs RTX 4090 — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-5090-vs-rtx-4090): 32 GB at $0.439/hr vs 24 GB at $0.327/hr
- [RTX 4090 vs RTX 3090 — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-4090-vs-rtx-3090): 24 GB at $0.327/hr vs 24 GB at $0.108/hr
- [RTX 4090 vs A100 PCIE — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-4090-vs-a100-pcie): 24 GB at $0.327/hr vs 80 GB at $0.374/hr
- [RTX 4090 vs L40S — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-4090-vs-l40s): 24 GB at $0.327/hr vs 48 GB at $0.514/hr
- [L40S vs A100 PCIE — specs, price per hour, which to rent](https://powergpu.ai/compare/l40s-vs-a100-pcie): 48 GB at $0.514/hr vs 80 GB at $0.374/hr
- [L40S vs H100 PCIE — specs, price per hour, which to rent](https://powergpu.ai/compare/l40s-vs-h100-pcie): 48 GB at $0.514/hr vs 80 GB at $1.867/hr
- [RTX 5090 vs L40S — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-5090-vs-l40s): 32 GB at $0.439/hr vs 48 GB at $0.514/hr
- [RTX 5090 vs A100 SXM4 — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-5090-vs-a100-sxm4): 32 GB at $0.439/hr vs 80 GB at $0.560/hr
- [RTX PRO 6000 WS vs RTX 6000Ada — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-pro-6000-ws-vs-rtx-6000ada): 96 GB at $1.040/hr vs 48 GB at $0.467/hr
- [RTX PRO 6000 WS vs H100 PCIE — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-pro-6000-ws-vs-h100-pcie): 96 GB at $1.040/hr vs 80 GB at $1.867/hr
- [RTX PRO 6000 WS vs RTX 5090 — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-pro-6000-ws-vs-rtx-5090): 96 GB at $1.040/hr vs 32 GB at $0.439/hr
- [RTX 6000Ada vs RTX A6000 — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-6000ada-vs-rtx-a6000): 48 GB at $0.467/hr vs 48 GB at $0.281/hr
- [RTX A6000 vs RTX 4090 — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-a6000-vs-rtx-4090): 48 GB at $0.281/hr vs 24 GB at $0.327/hr
- [L4 vs Tesla T4 — specs, price per hour, which to rent](https://powergpu.ai/compare/l4-vs-tesla-t4): 24 GB at $0.225/hr vs 16 GB at $0.103/hr
- [L4 vs A10 — specs, price per hour, which to rent](https://powergpu.ai/compare/l4-vs-a10): 24 GB at $0.225/hr vs 24 GB at $0.168/hr
- [Tesla V100 vs RTX 3090 — specs, price per hour, which to rent](https://powergpu.ai/compare/tesla-v100-vs-rtx-3090): 16 GB at $0.130/hr vs 24 GB at $0.108/hr
- [RTX 3090 vs RTX 3060 — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-3090-vs-rtx-3060): 24 GB at $0.108/hr vs 12 GB at $0.042/hr
- [RTX 5080 vs RTX 4090 — specs, price per hour, which to rent](https://powergpu.ai/compare/rtx-5080-vs-rtx-4090): 16 GB at $0.186/hr vs 24 GB at $0.327/hr
- [H100 NVL vs H100 SXM — specs, price per hour, which to rent](https://powergpu.ai/compare/h100-nvl-vs-h100-sxm): 80 GB at $1.811/hr vs 80 GB at $1.428/hr
- [H200 vs H200 NVL — specs, price per hour, which to rent](https://powergpu.ai/compare/h200-vs-h200-nvl): 141 GB at $2.791/hr vs 141 GB at $2.650/hr
- [B200 vs B300 — specs, price per hour, which to rent](https://powergpu.ai/compare/b200-vs-b300): 192 GB at $5.425/hr vs 288 GB at $6.737/hr

## Guides

- [Cloud GPU guides](https://powergpu.ai/guides): Costs, hardware choices, hands-on walkthroughs with live prices
- [Cloud GPU pricing explained (2026): on-demand vs spot vs reserved](https://powergpu.ai/guides/cloud-gpu-pricing-explained): What actually drives cloud GPU prices in 2026, how marketplaces and hyperscalers differ, and when each billing mode saves you money.
- [LLM VRAM requirements: how much GPU memory for 7B–405B models](https://powergpu.ai/guides/llm-vram-requirements): Complete sizing tables for running and training LLMs — FP16, INT8 and 4-bit, with KV-cache math and the cheapest GPU that fits each model.
- [H100 vs H200 vs B200 (2026): specs, price per hour, which to rent](https://powergpu.ai/guides/h100-vs-h200-vs-b200): Memory, bandwidth and real rental economics of NVIDIA's three datacenter flagships — and when the older card is the better deal.
- [RTX 5090 vs RTX 4090 for AI (2026): benchmarks, VRAM, rental cost](https://powergpu.ai/guides/rtx-4090-vs-rtx-5090): Specs, VRAM, real throughput differences and cost per run for inference, fine-tuning and image generation on both consumer flagships.
- [How to fine-tune Llama 3.1 8B with QLoRA on a single GPU](https://powergpu.ai/guides/fine-tune-llm-qlora): A complete, copy-pasteable walkthrough: dataset to merged weights in about an hour on a single 24 GB card, with axolotl.
- [How to deploy vLLM on a cloud GPU: an OpenAI-compatible endpoint](https://powergpu.ai/guides/serve-llm-vllm): Deploy vLLM on a rented GPU, pick the right card for your model size, benchmark tokens per second and put a price on every million tokens.
- [Cheapest cloud GPU for Stable Diffusion & Flux in 2026 ($/image)](https://powergpu.ai/guides/cheapest-gpu-for-stable-diffusion): What SDXL and Flux actually need, images-per-dollar on eight rentable GPUs, and where paying more per hour costs less per image.
- [Blender cloud rendering on GPUs: setup, cost per frame, pitfalls](https://powergpu.ai/guides/blender-cloud-rendering): Render Cycles scenes on rented RTX hardware: headless setup, per-frame cost math, and the mistakes that quietly triple a render bill.
- [How much does it cost to rent an H100? Per-hour math for 2026](https://powergpu.ai/guides/h100-rental-cost-per-hour): H100 SXM, PCIe and NVL rental prices per hour and per month, what marketplaces and hyperscalers charge, the rent-vs-buy break-even, and five ways to pay less.
- [Best cloud GPU for LLM inference in 2026, by model size](https://powergpu.ai/guides/best-cloud-gpu-for-llm-inference): From 8B to 405B: the cheapest rentable card that holds each model, indicative tokens per second with vLLM, and the cost per million tokens on today's sheet.
- [How to run ComfyUI on a cloud GPU: setup, models, cost per image](https://powergpu.ai/guides/run-comfyui-on-a-cloud-gpu): Deploy ComfyUI on a rented RTX 4090 or 5090 in 30 seconds, keep checkpoints on a volume, run workflows headless through the API, and know what each image costs.
- [How to rent a GPU with crypto and no KYC (2026)](https://powergpu.ai/guides/rent-gpu-with-crypto-no-kyc): Renting cloud GPUs with USDT, Bitcoin or Monero and no identity check: how top-ups work, which coin to use, what data is kept, and the step-by-step deploy.
- [A100 vs H100 for fine-tuning (2026): cost per run, not per hour](https://powergpu.ai/guides/a100-vs-h100-for-fine-tuning): Same 80 GB, 2.7× the hourly price: when an H100 finishes a LoRA, QLoRA or full fine-tune fast enough to beat the A100 on total cost — live prices, break-even rule.
- [Cheapest cloud GPU for Ollama (2026): 8B to 70B models, by the hour](https://powergpu.ai/guides/cheapest-gpu-for-ollama): Which rented card runs each Ollama model size at Q4, indicative tokens per second per card, and what an always-on private assistant costs per month.
- [Wan 2.x video generation: which GPU, minutes per clip, cost per clip](https://powergpu.ai/guides/wan-video-generation-gpu): VRAM needs for Wan 2.1 and 2.2 (1.3B, 5B, 14B), realistic minutes per five-second clip on RTX 4090, 5090, L40S, H100 and H200, and the price of a batch night.
- [Best GPU for Whisper transcription (2026): speed and cost per audio hour](https://powergpu.ai/guides/best-gpu-for-whisper): faster-whisper large-v3 throughput on T4, L4, RTX 3060, RTX 4090 and H100 — and the only number that matters: cents per hour of audio transcribed.

## Provider comparisons

- [GPU cloud alternatives — 34 providers compared with PowerGPU](https://powergpu.ai/alternatives): Public list prices with their check date, billing, payment and identity rules, migration map
- [RunPod alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/runpod): Developer GPU cloud: Pods, Serverless endpoints, Community and Secure tiers
- [Lambda Labs alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/lambda): ML-focused cloud: on-demand instances, 1-Click Clusters, Lambda Stack images
- [CoreWeave alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/coreweave): Enterprise AI hyperscaler: Kubernetes-native, 8-GPU HGX nodes, reserved-first
- [Paperspace alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/paperspace): Now part of DigitalOcean: GPU Droplets, Paperspace machines and Gradient notebooks
- [TensorDock alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/tensordock): GPU marketplace: independent hosts set prices across 100+ locations
- [Vultr GPU alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/vultr): General cloud with GPU instances and bare metal in 32 regions
- [Hyperstack alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/hyperstack): European GPU cloud with per-minute billing and NVLink options
- [SaladCloud alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/salad): Distributed consumer GPUs on volunteer PCs — the lowest hourly prices, with the trade-offs
- [io.net alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/io-net): Decentralized GPU network: clusters aggregated from independent suppliers, paid in crypto
- [Modal alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/modal): Python-native serverless: functions on GPUs, per-second, scale to zero
- [Jarvislabs alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/jarvislabs): India- and EU-based GPU cloud with per-minute billing and pause/resume
- [AWS GPU alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/aws): Amazon EC2 GPU instances: P5 (H100), P4d (A100), G6e (L40S), G5 (A10G), G4dn (T4)
- [Google Cloud GPU alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/google-cloud): Compute Engine A3 (H100), A2 (A100), G2 (L4) machine types and Vertex AI
- [Azure GPU alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/azure): ND H100 v5, NC A100 v4 and NC T4 virtual machines with enterprise agreements
- [Akash Network alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/akash): Decentralized GPU marketplace: permissionless providers bid, settled on-chain in AKT or USDC
- [Crusoe alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/crusoe): AI cloud built on stranded and flared energy: H100, H200, A100, L40S on a published rate card
- [Cudo Compute alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/cudo-compute): NVIDIA Cloud Partner marketplace; live pricing page is quote-only, not self-serve
- [DataCrunch alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/datacrunch): Renamed Verda in Nov. 2025: own Nordic datacenters, H100 through B300, no crypto payment
- [FluidStack alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/fluidstack): Neocloud turned AI-factory builder: dedicated, multi-year contracts for a handful of large labs
- [Genesis Cloud alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/genesis-cloud): Munich GPU cloud on Nordic capacity; GmbH in formal liquidation since August 2025
- [Hetzner GPU alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/hetzner): Dedicated GPU servers by the month (or hour): RTX PRO 4000/6000 Blackwell, Germany/Finland
- [Hyperbolic alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/hyperbolic): GPU marketplace and inference API: weekly-refreshed on-demand rates, reserved and private-cloud tiers
- [Latitude.sh alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/latitude-sh): Bare metal GPU by the hour or the month: H100, RTX PRO 6000, HGX B300 across the Americas, Europe and Asia
- [Lightning AI alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/lightning-ai): Cloud IDE with built-in GPUs: Studios, per-hour billing, monthly free credits
- [Massed Compute alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/massed-compute): NVIDIA Preferred Partner: on-demand GPU VMs, bare metal, US-only, per-minute billing
- [Nebius AI Cloud alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/nebius): Ex-Yandex hyperscaler-scale cloud: owned Finland/France datacenters, H100 through B300
- [Novita AI alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/novita): GPU instances, serverless endpoints and hosted model APIs, strong APAC reach
- [Oracle Cloud GPU alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/oracle-cloud): OCI bare metal and VMs: H100, H200, A100, L40S, A10, preemptible at 50% off
- [OVHcloud GPU alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/ovhcloud): European sovereign cloud: hourly Public Cloud GPUs, Scale/HGR-AI bare metal by quote
- [Prime Intellect alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/prime-intellect): Multi-provider compute exchange: aggregated on-demand GPUs, multi-node clusters, decentralized training runs
- [Replicate alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/replicate): Hosted models via API, Cog containers, billed per second of GPU time
- [Scaleway GPU alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/scaleway): European GPU cloud: H100 SXM, L40S, L4 by the hour, free egress, Paris/Amsterdam/Warsaw
- [Together AI alternative — prices vs PowerGPU](https://powergpu.ai/alternatives/together-ai): Inference and fine-tuning API, plus dedicated GPU Clusters on H100, H200, B200

## Templates

- [37 one-click GPU templates](https://powergpu.ai/templates): Official images: PyTorch, vLLM, ComfyUI, Ollama, Hugging Face, Ubuntu VMs…
- [Run NVIDIA CUDA on a cloud GPU](https://powergpu.ai/templates/nvidia-cuda): The bare CUDA base image — drivers, toolkit, SSH and Jupyter. Build your own stack on top.
- [Run PyTorch on a cloud GPU](https://powergpu.ai/templates/pytorch): The training and research default: PyTorch 2.6, cuDNN, JupyterLab and SSH out of the box.
- [Run TensorFlow CUDA on a cloud GPU](https://powergpu.ai/templates/tensorflow-cuda): TensorFlow with GPU support, Keras and TensorBoard on a mapped port.
- [Run PyTorch NGC on a cloud GPU](https://powergpu.ai/templates/pytorch-ngc): NVIDIA's NGC-optimised PyTorch build — tuned kernels for the datacenter cards.
- [Run NVIDIA RAPIDS on a cloud GPU](https://powergpu.ai/templates/nvidia-rapids): GPU-accelerated data science — cuDF, cuML, cuGraph in a notebook.
- [Run All-in-One App Studio on a cloud GPU](https://powergpu.ai/templates/all-in-one-app-studio): A launcher bundling the most-used AI apps behind one desktop — pick and run.
- [Run vLLM on a cloud GPU](https://powergpu.ai/templates/vllm): Production LLM serving with PagedAttention — an OpenAI-compatible endpoint from any HF model.
- [Run vLLM Omni on a cloud GPU](https://powergpu.ai/templates/vllm-omni): vLLM extended for multimodal models — vision and audio inputs on the same fast server.
- [Run SGLang on a cloud GPU](https://powergpu.ai/templates/sglang): High-throughput serving with RadixAttention — excels at structured and agentic workloads.
- [Run Llama.cpp on a cloud GPU](https://powergpu.ai/templates/llama-cpp): GGUF inference with a built-in server — the lightest way to run quantized models.
- [Run Ollama on a cloud GPU](https://powergpu.ai/templates/ollama): Pull and run quantized models in one command, with a clean REST API.
- [Run Open WebUI (Ollama) on a cloud GPU](https://powergpu.ai/templates/open-webui-ollama): A full chat UI over Ollama — conversations, RAG and model management in the browser.
- [Run Oobabooga Text Gen on a cloud GPU](https://powergpu.ai/templates/oobabooga-text-gen): The text-generation WebUI & API — load GPTQ, EXL2 or GGUF with extensions.
- [Run HuggingFace TGI on a cloud GPU](https://powergpu.ai/templates/huggingface-tgi): Hugging Face Text Generation Inference — battle-tested production serving with an API.
- [Run Langflow (Ollama) on a cloud GPU](https://powergpu.ai/templates/langflow-ollama): Visually build LLM pipelines and agents on top of a local Ollama backend.
- [Run Axolotl — Fine Tuning on a cloud GPU](https://powergpu.ai/templates/axolotl-fine-tuning): Fine-tune Llama, Qwen or Mistral from one YAML — LoRA, QLoRA and multi-GPU FSDP.
- [Run Unsloth Studio on a cloud GPU](https://powergpu.ai/templates/unsloth-studio): 2× faster, lower-VRAM fine-tuning — Unsloth kernels with a notebook workflow.
- [Run Kohya's GUI on a cloud GPU](https://powergpu.ai/templates/kohya-s-gui): Train SDXL and Flux image LoRAs in a browser GUI — dataset in, .safetensors out.
- [Run Flux Gym on a cloud GPU](https://powergpu.ai/templates/flux-gym): A dead-simple UI for training Flux LoRAs, built on the Kohya scripts.
- [Run Ostris AI Toolkit on a cloud GPU](https://powergpu.ai/templates/ostris-ai-toolkit): The AI-Toolkit trainer for Flux and diffusion models, with a web UI.
- [Run ComfyUI on a cloud GPU](https://powergpu.ai/templates/comfyui): Node-based image and video workflows with ComfyUI-Manager pre-installed.
- [Run InvokeAI on a cloud GPU](https://powergpu.ai/templates/invokeai): A polished Stable Diffusion studio — canvas, layers and workflow nodes.
- [Run SD WebUI Forge on a cloud GPU](https://powergpu.ai/templates/sd-webui-forge): The Forge fork of the SD WebUI — faster attention, lower VRAM, same extensions.
- [Run SD WebUI A1111 on a cloud GPU](https://powergpu.ai/templates/sd-webui-a1111): The classic AUTOMATIC1111 Stable Diffusion WebUI with the full extension ecosystem.
- [Run Fooocus on a cloud GPU](https://powergpu.ai/templates/fooocus): Prompt-and-go image generation — Midjourney-style simplicity on SDXL.
- [Run SwarmUI on a cloud GPU](https://powergpu.ai/templates/swarmui): A ComfyUI-backed UI that scales generation across multiple GPUs.
- [Run Wan2GP on a cloud GPU](https://powergpu.ai/templates/wan2gp): Run Wan and other video-generation models on modest VRAM, with a simple UI.
- [Run Open-Sora on a cloud GPU](https://powergpu.ai/templates/open-sora): The open text-to-video model, ready to generate and fine-tune.
- [Run Whisper WebUI & API on a cloud GPU](https://powergpu.ai/templates/whisper-webui-api): Batch-transcribe audio with faster-whisper behind a UI and a REST endpoint.
- [Run Voicebox TTS on a cloud GPU](https://powergpu.ai/templates/voicebox-tts): Text-to-speech and voice cloning with a web interface and API.
- [Run ACE Step 1.5 on a cloud GPU](https://powergpu.ai/templates/ace-step-1-5): The ACE-Step music generation model with a UI and API.
- [Run Linux Desktop on a cloud GPU](https://powergpu.ai/templates/linux-desktop): A full GPU Linux desktop over the browser — XFCE, VNC/RDP, run GUI apps.
- [Run Pinokio on a cloud GPU](https://powergpu.ai/templates/pinokio): The Pinokio 1-click app browser on a desktop — install AI apps with a click.
- [Run Ubuntu 22.04 VM on a cloud GPU](https://powergpu.ai/templates/ubuntu-22-04-vm): A full KVM virtual machine — your own kernel, root, systemd, any stack.
- [Run Ubuntu Desktop VM on a cloud GPU](https://powergpu.ai/templates/ubuntu-desktop-vm): The Ubuntu VM with a GNOME desktop over VNC/RDP — a full graphical workstation.
- [Run Unreal Pixel Streaming on a cloud GPU](https://powergpu.ai/templates/unreal-pixel-streaming): Stream an Unreal Engine app to the browser — pixel streaming on a cloud GPU.
- [Run Hashcat CUDA on a cloud GPU](https://powergpu.ai/templates/hashcat-cuda): GPU password recovery for authorised security testing — CUDA-accelerated Hashcat.

## Documentation & developers

- [Documentation](https://powergpu.ai/docs)
- [Quickstart — account to running GPU in five minutes](https://powergpu.ai/docs/quickstart)
- [Billing & credits](https://powergpu.ai/docs/billing)
- [Instance lifecycle](https://powergpu.ai/docs/instances)
- [Templates & images](https://powergpu.ai/docs/templates)
- [Volumes & storage](https://powergpu.ai/docs/volumes)
- [Networking & ports](https://powergpu.ai/docs/networking)
- [API keys](https://powergpu.ai/docs/api-keys)
- [Quotas & limits](https://powergpu.ai/docs/limits)
- [FAQ](https://powergpu.ai/docs/faq)
- [MCP server — connect AI agents to PowerGPU](https://powergpu.ai/docs/mcp)
- [REST API reference](https://powergpu.ai/api): Public pricing and offers endpoints (no key), authenticated instance lifecycle
- [CLI reference](https://powergpu.ai/cli)
- [Python SDK reference](https://powergpu.ai/sdk)

## Company, products & legal

- [PowerGPU — cloud GPU rental at fixed prices, 30% below market](https://powergpu.ai/): Home: 80 NVIDIA GPUs, price rule, how it works, FAQ
- [Enterprise GPU cloud — reserved fleets, clusters, volume pricing](https://powergpu.ai/enterprise)
- [About PowerGPU — the fixed-price model](https://powergpu.ai/about)
- [Security — isolation, encryption, verified datacenters](https://powergpu.ai/security)
- [GPU cloud use cases](https://powergpu.ai/use-cases): Training, inference, fine-tuning, image and video generation, rendering, vision, HPC
- [LLM training GPUs — recommended cards and costs](https://powergpu.ai/use-cases/llm-training): Pre-train and continue-train transformer models on HBM GPUs with NVLink and InfiniBand clusters.
- [LLM inference GPUs — recommended cards and costs](https://powergpu.ai/use-cases/llm-inference): Serve 7B–70B models with vLLM or TensorRT-LLM at the lowest $ per million tokens.
- [Fine-tuning GPUs — recommended cards and costs](https://powergpu.ai/use-cases/fine-tuning): LoRA, QLoRA and full fine-tunes — from a single 24 GB card to 8× A100 nodes.
- [Image generation GPUs — recommended cards and costs](https://powergpu.ai/use-cases/image-generation): SDXL, Flux and ComfyUI pipelines — interactive sessions or thousand-image batch runs.
- [Video generation GPUs — recommended cards and costs](https://powergpu.ai/use-cases/video-generation): Wan, HunyuanVideo and image-to-video models need VRAM headroom — rent it by the second.
- [3D rendering GPUs — recommended cards and costs](https://powergpu.ai/use-cases/rendering): Blender, Octane, Redshift and V-Ray on RTX hardware — per-second billing fits render farms perfectly.
- [Computer vision GPUs — recommended cards and costs](https://powergpu.ai/use-cases/computer-vision): Train YOLO and detection models, run batch inference over image and video archives.
- [Scientific computing GPUs — recommended cards and costs](https://powergpu.ai/use-cases/hpc): FP64, huge memory bandwidth and MPI clusters for simulation, genomics and quantitative research.
- [On-demand GPUs — Fixed price, guaranteed resources, per-second billing](https://powergpu.ai/products/on-demand)
- [Interruptible — Half the on-demand price for fault-tolerant jobs](https://powergpu.ai/products/interruptible)
- [Reserved — Commit 3+ months, save another 35%](https://powergpu.ai/products/reserved)
- [Serverless — Autoscaling endpoints for vLLM, ComfyUI and your images](https://powergpu.ai/products/serverless)
- [Clusters — Multi-node InfiniBand/NVLink for distributed training](https://powergpu.ai/products/clusters)
- [Virtual machines — Full VMs with root, custom kernels and GUI images](https://powergpu.ai/products/vms)
- [Volumes — NVMe network storage at $0.08/GB/mo](https://powergpu.ai/products/volumes)
- [Terms of service](https://powergpu.ai/legal/terms)
- [Privacy policy](https://powergpu.ai/legal/privacy)
- [Acceptable use policy](https://powergpu.ai/legal/aup)
- [Service level agreement](https://powergpu.ai/legal/sla)

## Optional

- [Full site content in Markdown](https://powergpu.ai/llms-full.txt): every page above, converted
- [Markdown for any page](https://powergpu.ai/pricing.md): append .md to a page URL, or send Accept: text/markdown
- [XML sitemap index](https://powergpu.ai/sitemap.xml)
- [RSS feed of guides](https://powergpu.ai/feed)
- [Service status JSON](https://powergpu.ai/status.json)