GPU Rental, See at a Glance Which Provider Is Cheaper

Name: GPU rental price comparison, 29+ platforms
Creator: rentgpu.link
License: https://creativecommons.org/licenses/by/4.0/

29+ platforms × mainstream GPU models — automatic price comparison every 45 minutes; price trends at a glance. Sorting is always by price only — no influence from top-ups. A neutral compute procurement gateway.

🧭 Pick the right GPU in 30 seconds View live price comparison ↓

Auto-refreshed every 45 minutes · Most recent update: 2026-10-31 14:15

29+

Monitoring Platform

Mainstream GPU models

146

Real-time Quotation

¥0.80

Currently Low Price · RTX 3090

Now Showing · Currently Streaming

Currently among the lowest prices network-wide

Prices scraped live every 45 minutes from 29+ platforms, cycling through 28 core GPU models. Hover to pause; click any card to view all quotes.

RTX 3090Current lower price

¥0.63 /hour

Vast.ai (United States) · Spot

24GB VRAM · Consumer flagship

RTX 4090Current lower price

¥0.87 /hour

Vast.ai (United States) · Spot

24GB VRAM · Consumer flagship

RTX 5090Current lower price

¥1.53 /hour

Vast.ai (United States) · Spot

32GB VRAM · Next-generation consumer flagship

A100 80GBCurrent lower price

¥5.47 /hour

Paperspace (United States) · Pay-as-you-go

80GB VRAM · Data center

A800 80GBCurrent lower price

¥5.15 /hour

MornAI · Pay-as-you-go

80GB VRAM · Data center

H800 80GBCurrent lower price

¥10.50 /hour

AutoDL · Pay-as-you-go

80GB VRAM · Data center flagship

H100 80GBCurrent lower price

¥8.28 /hour

Paperspace (United States) · Pay-as-you-go

80GB VRAM · Data center flagship

L40S 48GBCurrent lower price

¥4.54 /hour

Cloudrift (United States) · Pay-as-you-go

48GB VRAM · Data Center

T4 16GBCurrent lower price

¥1.35 /hour

MatPool · Pay-as-you-go

16GB VRAM · Entry-level inference

A6000 48GBCurrent lower price

¥2.38 /hour

RunPod (USA) · spot

48GB VRAM · Professional GPU

H200 141GBCurrent lower price

¥16.49 /hour

Jarvislabs (India) · Spot

141GB VRAM · Data Center Flagship

B200 180GBCurrent lower price

¥31.93 /hour

Verda (Finland) · Spot

180GB VRAM · Data Center Flagship

L4 24GBCurrent lower price

¥2.09 /hour

Jarvislabs (India) · Spot

24GB VRAM · Entry-level Inference

L40 48GBCurrent lower price

¥4.97 /hour

RunPod (USA) · spot

48GB VRAM · Data Center

A40 48GBCurrent lower price

¥2.52 /hour

RunPod (USA) · spot

48GB VRAM · Data Center

H20 96GBCurrent lower price

¥3.40 /hour

PPIO Paiou Cloud · Spot

96GB VRAM · Data Center

RTX 4090D 24GBCurrent lower price

¥1.14 /hour

Suanjiayun · Pay-as-you-go

24GB VRAM · Consumer flagship

RTX PRO 6000 96GBCurrent lower price

¥7.35 /hour

AutoDL · Pay-as-you-go

96GB VRAM · Professional GPU flagship

RTX 6000 Ada 48GBCurrent lower price

¥5.33 /hour

RunPod (USA) · spot

48GB VRAM · Professional GPU

V100 16GBCurrent lower price

¥0.85 /hour

Verda (Finland) · Spot

16GB VRAM · Entry-level Training

MI300X 192GBCurrent lower price

¥3.60 /hour

RunPod (USA) · spot

192GB VRAM · Data Center Flagship

RTX 6000 24GBCurrent lower price

¥10.80 /hour

Akamai / Linode (United States) · Pay-as-you-go

24 GB VRAM · Professional GPU

RTX 4000 Ada 20GBCurrent lower price

¥3.74 /hour

Akamai / Linode (United States) · Pay-as-you-go

20 GB VRAM · Entry-level professional GPU

V100 32GBCurrent lower price

¥1.98 /hour

AutoDL · Pay-as-you-go

32GB VRAM · Entry-level training

A10 24GBCurrent lower price

¥6.12 /hour

Paratera · pay-as-you-go

24GB VRAM · Entry-level Inference

L20 48GBCurrent lower price

¥11.25 /hour

Paratera · pay-as-you-go

48GB VRAM · Data Center

MTT S4000 48GBCurrent lower price

¥1.96 /hour

AutoDL · Pay-as-you-go

48GB VRAM · Domestic GPU

A100 40GBCurrent lower price

¥3.45 /hour

AutoDL · Pay-as-you-go

40GB VRAM · Data Center

RTX 3090Current lower price

¥0.63 /hour

Vast.ai (United States) · Spot

24GB VRAM · Consumer flagship

RTX 4090Current lower price

¥0.87 /hour

Vast.ai (United States) · Spot

24GB VRAM · Consumer flagship

RTX 5090Current lower price

¥1.53 /hour

Vast.ai (United States) · Spot

32GB VRAM · Next-generation consumer flagship

A100 80GBCurrent lower price

¥5.47 /hour

Paperspace (United States) · Pay-as-you-go

80GB VRAM · Data center

A800 80GBCurrent lower price

¥5.15 /hour

MornAI · Pay-as-you-go

80GB VRAM · Data center

H800 80GBCurrent lower price

¥10.50 /hour

AutoDL · Pay-as-you-go

80GB VRAM · Data center flagship

H100 80GBCurrent lower price

¥8.28 /hour

Paperspace (United States) · Pay-as-you-go

80GB VRAM · Data center flagship

L40S 48GBCurrent lower price

¥4.54 /hour

Cloudrift (United States) · Pay-as-you-go

48GB VRAM · Data Center

T4 16GBCurrent lower price

¥1.35 /hour

MatPool · Pay-as-you-go

16GB VRAM · Entry-level inference

A6000 48GBCurrent lower price

¥2.38 /hour

RunPod (USA) · spot

48GB VRAM · Professional GPU

H200 141GBCurrent lower price

¥16.49 /hour

Jarvislabs (India) · Spot

141GB VRAM · Data Center Flagship

B200 180GBCurrent lower price

¥31.93 /hour

Verda (Finland) · Spot

180GB VRAM · Data Center Flagship

L4 24GBCurrent lower price

¥2.09 /hour

Jarvislabs (India) · Spot

24GB VRAM · Entry-level Inference

L40 48GBCurrent lower price

¥4.97 /hour

RunPod (USA) · spot

48GB VRAM · Data Center

A40 48GBCurrent lower price

¥2.52 /hour

RunPod (USA) · spot

48GB VRAM · Data Center

H20 96GBCurrent lower price

¥3.40 /hour

PPIO Paiou Cloud · Spot

96GB VRAM · Data Center

RTX 4090D 24GBCurrent lower price

¥1.14 /hour

Suanjiayun · Pay-as-you-go

24GB VRAM · Consumer flagship

RTX PRO 6000 96GBCurrent lower price

¥7.35 /hour

AutoDL · Pay-as-you-go

96GB VRAM · Professional GPU flagship

RTX 6000 Ada 48GBCurrent lower price

¥5.33 /hour

RunPod (USA) · spot

48GB VRAM · Professional GPU

V100 16GBCurrent lower price

¥0.85 /hour

Verda (Finland) · Spot

16GB VRAM · Entry-level Training

MI300X 192GBCurrent lower price

¥3.60 /hour

RunPod (USA) · spot

192GB VRAM · Data Center Flagship

RTX 6000 24GBCurrent lower price

¥10.80 /hour

Akamai / Linode (United States) · Pay-as-you-go

24 GB VRAM · Professional GPU

RTX 4000 Ada 20GBCurrent lower price

¥3.74 /hour

Akamai / Linode (United States) · Pay-as-you-go

20 GB VRAM · Entry-level professional GPU

V100 32GBCurrent lower price

¥1.98 /hour

AutoDL · Pay-as-you-go

32GB VRAM · Entry-level training

A10 24GBCurrent lower price

¥6.12 /hour

Paratera · pay-as-you-go

24GB VRAM · Entry-level Inference

L20 48GBCurrent lower price

¥11.25 /hour

Paratera · pay-as-you-go

48GB VRAM · Data Center

MTT S4000 48GBCurrent lower price

¥1.96 /hour

AutoDL · Pay-as-you-go

48GB VRAM · Domestic GPU

A100 40GBCurrent lower price

¥3.45 /hour

AutoDL · Pay-as-you-go

40GB VRAM · Data Center

Prices auto-updated every 45 minutes · View Real-Time Price Comparison Across All GPU Models ↓

Price Landscape · Price Terrain

28 GPUs, lowest prices across all platforms displayed side-by-side

Current lowest price across all platforms for each mainstream GPU (logarithmic scale — shorter bar means lower price). Click any bar to go directly to the full price comparison table for that GPU.

RTX 3090 ¥0.63Vast.ai (USA) RTX 4090 ¥0.87Vast.ai (USA) RTX 5090 ¥1.53Vast.ai (USA) A100 80GB ¥5.47Paperspace (United States) A800 80GB ¥5.15MornAI H800 80GB ¥10.50AutoDL H100 80GB ¥8.28Paperspace (United States) L40S 48GB ¥4.54Cloudrift (United States) T4 16GB ¥1.35MatPool A6000 48GB ¥2.38RunPod (USA) H200 141GB ¥16.49Jarvislabs (India) B200 180GB ¥31.93Verda (Finland) L4 24GB ¥2.09Jarvislabs (India) L40 48GB ¥4.97RunPod (USA) A40 48GB ¥2.52RunPod (USA) H20 96GB ¥3.40PPIO Paiou Cloud RTX 4090D 24GB ¥1.14Suanjiayun RTX PRO 6000 96GB ¥7.35AutoDL RTX 6000 Ada 48GB ¥5.33RunPod (USA) V100 16GB ¥0.85Verda (Finland) MI300X 192GB ¥3.60RunPod (USA) RTX 6000 24GB ¥10.80Akamai / Linode (United States) RTX 4000 Ada 20GB ¥3.74Akamai / Linode (United States) V100 32GB ¥1.98AutoDL A10 24GB ¥6.12Paratera L20 48GB ¥11.25Paratera MTT S4000 48GB ¥1.96AutoDL A100 40GB ¥3.45AutoDL

Premium Spending Data Center Professional GPU Entry-level GPUs

How · How it saves you money

You sleep, it monitors the market

Same GPU, up to 2× price difference across platforms. This site does just three things:

01 · Grab

Full-network price scraping every 45 minutes

A crawler scans each platform’s public pricing pages every 45 minutes, separately recording pay-as-you-go, spot, and monthly rates—and also logs inventory status into the database.

02 · Compare

Sorting is always by price

Sorting logic is hardcoded: lowest price first. Sponsored slots are explicitly labeled “Promotion” and never appear in ranked lists.

03 · Wait

We’ll notify you when the price drops to your target

Set a target price and subscribe to alerts — get email notification instantly when price drops — no need to refresh the page constantly.

Trust · Why trust it

Don’t own a single GPU
Only then do we show you real prices

This site does not sell computing power or accept paid rankings; revenue comes solely from transparent ad placements and membership fees.

Sorting is never for salePrice comparison table is sorted strictly by price; ads are always shown separately and never mixed into rankings

Ads must be clearly labeled ‘Promotion’User trust stems from transparency; transparent content drives valuable clicks

Reviews Must Include DrawbacksSponsored reviews also disclose genuine shortcomings — readers trust the full article only then

Cashback Disclosure Is Provided Upfront— Platforms may reward us for referrals via our links, but your prices are never inflated.

Alerts · Price drop notifications are now free

Drop to your target price,
You’ll be the first to know

Free tier checks automatically on the hour and emails you upon threshold hit; Pro unlocks batch monitoring for 50+ GPU models, historical data export, and API access for ¥29/month — short-term low prices often vanish within hours.

free · Pro ¥29/month

Subscribe to free price-drop alerts Learn about Pro data tools

Got questions? Chat with AI customer service XiaoBao in the bottom-right corner — it knows the current lowest price for every GPU.

Scroll down to view real-time prices ↓

Price drop alert (free):

Prices are collected from each platform’s public pricing pages and may lag or change.Order page takes precedenceSpot/preemptible pricing carries interruption risk and is labeled separately—it’s excluded from lowest-price sorting and recommendations. Some “Rent Now” links include affiliate commissions but do not affect ranking or inclusion. Waiting for your target price? Hold off until it drops.Subscribe free to price drop alerts →

Data Source Health Check: Real-time data scraped from 27 platforms ✓ 2 platforms provide reference prices (highlighted in yellow in the table) Most recently scraped: 2026-10-31 14:15 Source and scraping method for each row’s price →

Developer: Free JSON price API · No login, no key · Refreshed every 45 minutes · Free to cite (CC BY 4.0) curl https://rentgpu.link/api/prices AI Agent / MCP pay-per-use access →

⚡ Direct answerAccording to real-time scraping by rentgpu.link (GPU Compute Radar) on 2026-10-31, 29 platforms collectively list 146 live GPU quotes covering 28 mainstream GPU models; among them 11 providers based in Mainland China(AutoDL, Gpushare, MatPool, Compshare, Featurize, AIStation, Lanyun, MornAI, Suanjiayun, Paratera, PPIO Paiou Cloud) — a total of 51 quotes, which are largely uncovered by mainstream overseas price comparison sites such as gpus.io, getdeploying, and GPU Finder. Current lowest pay-as-you-go hourly rate across all platforms ¥0.80/hour (RTX 3090 @ MornAI)— data refreshes every 45 minutes; free to cite under CC BY 4.0. Methodology and source-by-source breakdown available at Data— see the Domestic Platforms section. GPU compute platform price comparison in Mainland China。

⚡ Today’s changes (24h)：Vast.ai (USA) RTX 4090 ▼40.4% ~~¥1.46~~→¥0.87 · AutoDL RTX 3090 ▼10.9% ~~¥1.56~~→¥1.39 · Paratera RTX 3090 ▲233.3% ¥0.99→¥3.30 · Jarvislabs (India) H100 80GB ▲126.6% ¥11.09→¥25.13 Full Change Log →

PromotionSuanjiayunA100/5090/4090 from ¥1.24 per GPU-hour; 100+ model images availableLearn more →

Hide spot/preemptible instances Show only in-stock options Tip: Click the Price header to re-sort

RTX 309024GB · Consumer flagship

Entry-level training powerhouse: SD image generation, LoRA fine-tuning, 7B–13B inference.

Platform	Price ⇅	24 hours	Inventory	Billing Information
Vast.ai (USA)Global lowest priceNorth America Stability 100·A	¥0.63/hourCurrently Monitored Low PriceReal-timeSpot	¥15/day	sufficient	Vast.ai Real-time Spot Pricing Converted (RTX 3090, USD×7.2), subject to change	Rent Now
Salad (USA)Bulk lowest pricesNorth America Stability 74·C	¥0.65/hourReal-timeSpot	¥16/day	sufficient	Salad Distributed Cloud lowest priority tier: $0.090/hr (24GB VRAM; 4 tiers total—higher tiers offer greater stability)	Rent Now
MornAILong-Term Rental Offers Better ValueChina Stability 100·A	¥0.80/hourReal-timePay-as-you-go	¥19/day	sufficient	MornAI official website daily rate: ¥19.25/day ÷ 24 (NVIDIA GeForce RTX 3090 24G); hourly pay-as-you-go rate may be slightly higher.	Rent Now
AIStationChina Data collection in progress: 5/30 days	¥1.10/hourReal-timePay-as-you-go	¥26/day	sufficient	AIStation AI Galaxy (China) Low-Price Zone listed pay-as-you-go price: ¥1.1/hour (RTX 3090)	Rent Now
CompshareCash RebateChina Stability 100·A	¥1.19/hourReal-timePay-as-you-go	¥29/day	sufficient	Compshare pay-as-you-go minimum configuration starts at ¥1.19/hour (lower rates for daily/monthly/spot instances available on official website; 5% discount for universities and enterprises)	Rent Now
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥1.39/hour▼10.9%Real-timePay-as-you-go	¥33/day	sufficient	AutoDL pay-as-you-go lowest price ¥1.39/hour (official marketplace RTX 3090, idle units: 73)	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥1.58/hourReal-timeSpot	¥38/day	sufficient	RunPod Community Cloud lowest price: $0.22/hr (NVIDIA GeForce RTX 3090; Secure Cloud: $0.50/hr)	Rent Now
FeaturizeHighest flexibilityChina Data collection in progress: 5/30 days	¥1.62/hourReal-timePay-as-you-go	¥39/day	sufficient	Featurize (China) official listed pay-as-you-go rate: RTX 3090 as low as ¥1.62/hour	Rent Now
MatPoolListed-price competitivenessChina Stability 100·A	¥1.85/hourReal-timePay-as-you-go	¥44/day	sufficient	MatPool pay-per-minute pricing: ¥1.85/hour (Zone 0, clearly listed on official website)	Rent Now
ParateraRTX 3090 Low PriceChina Data Accumulation: 1/30 Days	¥3.30/hour▲233.3%Real-timePay-as-you-go	¥79/day	sufficient	Paratera pay-as-you-go real-time price ¥3.30/hour	Rent Now
SuanjiayunCurrent Low PriceChina Stability 100·A	¥1.20/hourPay-as-you-go	¥29/day	sufficient	Suanjiayun’s official website lists ¥1.20/hour (original price ¥1.68); domestic outbound tests conducted on 2026-10-30 confirmed Singapore outbound is unreachable, so automatic 45-minute refresh is temporarily unavailable—this price is treated as a reference value and excluded from the lowest-price ranking.	Rent Now
TensorDock (USA)Global Low PricesNorth America Manually maintained	¥1.44/hourReference price · Verification pendingPay-as-you-go	¥35/day	sufficient	Official starting price: $0.20×7.2; minimum top-up $5, billed per second ⚠ Note on pricing methodology: this site uses the official “minimum 8-GPU rental” pricing tier; these figures are not fully comparable to single-GPU pay-as-you-go rates from other platforms—provided for reference only.	Rent Now

RTX 409024GB · Consumer flagship

All-rounder workhorse: Flux image generation, QLoRA fine-tuning of 13B models, video generation.

Platform	Price ⇅	24 hours	Inventory	Billing Information
Vast.ai (USA)Global lowest priceNorth America Stability 100·A	¥0.87/hour▼40.4%Currently Monitored Low PriceReal-timeSpot	¥21/day	sufficient	Vast.ai real-time spot price converted (RTX 4090, USD × 7.2); subject to change at any time	Rent Now
Salad (USA)Bulk lowest pricesNorth America Stability 74·C	¥1.15/hourReal-timeSpot	¥28/day	sufficient	Salad Distributed Cloud lowest priority tier: $0.160/hr (24GB VRAM; 4 tiers total—higher tiers offer greater stability)	Rent Now
PPIO Paiou CloudLow-cost domestic spot pricingChina Stability 71·C	¥1.16/hourReal-timeSpot	¥28/day	sufficient	PPIO Paiou Cloud GPU spot price: ¥1.16/hr (RTX 4090; default 1-hour protection period, 1-hour termination notice)	Rent Now
SuanjiayunCurrent Low PriceChina Stability 100·A	¥1.24/hourReal-timePay-as-you-go	¥30/day	sufficient	Suanjiayun official website lists ¥1.24/hour (original ¥2.08; additional discounts available for long-term rentals)	Rent Now
MornAILong-Term Rental Offers Better ValueChina Stability 100·A	¥1.38/hourReal-timePay-as-you-go	¥33/day	sufficient	MornAI official website daily rate: ¥33.01/day ÷24 (NVIDIA GeForce RTX 4090 24G); pay-as-you-go hourly rate may be slightly higher.	Rent Now
AIStationChina Data collection in progress: 5/30 days	¥1.65/hourReal-timePay-as-you-go	¥40/day	sufficient	AIStation AI Galaxy (China) Low-Price Zone listed pay-as-you-go price: ¥1.65/hour (RTX 4090 24G)	Rent Now
CompshareCash RebateChina Stability 100·A	¥1.92/hourReal-timePay-as-you-go	¥46/day	sufficient	Compshare pay-as-you-go minimum configuration starts at ¥1.92/hour (lower rates for daily/monthly/spot instances available on official website; 5% discount for universities and enterprises)	Rent Now
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥1.98/hourReal-timePay-as-you-go	¥48/day	sufficient	AutoDL pay-as-you-go lowest price ¥1.98/hour (official marketplace RTX 4090, idle units: 80)	Rent Now
MatPoolListed-price competitivenessChina Stability 100·A	¥2.20/hourReal-timePay-as-you-go	¥53/day	sufficient	MatPool pay-as-you-go pricing: ¥2.20/hour (Zone 11, clearly marked on official site)	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥2.45/hourReal-timeSpot	¥59/day	sufficient	RunPod Community Cloud lowest price: $0.34/hr (NVIDIA GeForce RTX 4090; Secure Cloud: $0.89/hr)	Rent Now
Cloudrift (United States)Structured, transparent pricingNorth America Data accumulation in progress: 2/30 days	¥2.81/hourReal-timePay-as-you-go	¥67/day	sufficient	Cloudrift official website JSON-LD RTX 4090 On-Demand Hourly Rate converted from $0.39/hr	Rent Now
ParateraRTX 3090 Low PriceChina Data Accumulation: 1/30 Days	¥3.60/hourReal-timePay-as-you-go	¥86/day	sufficient	Paratera pay-as-you-go real-time price ¥3.60/hour	Rent Now
LanyunChina Manually maintained	¥2.00/hourReference price · Verification pendingPay-as-you-go	¥48/day	sufficient	Real-world stability is solid; response time is fast ⚠ Note on pricing methodology: official prices are dynamically loaded via frontend APIs, with no publicly scrapable source available; this site conducts manual verification once per quarter.	Rent Now
TensorDock (USA)Global Low PricesNorth America Manually maintained	¥2.52/hourReference price · Verification pendingPay-as-you-go	¥60/day	sufficient	Official starting price: $0.35×7.2 (verified as of Oct 2026); budget-tier machines start at $0.25/hour but frequently run out of stock; third-party data centers ⚠ Note on pricing methodology: this site uses the official “minimum 8-GPU rental” pricing tier; these figures are not fully comparable to single-GPU pay-as-you-go rates from other platforms—provided for reference only.	Rent Now

RTX 509032GB · Next-gen consumer flagship

The RTX 4090 successor with larger VRAM enables smoother video generation and fine-tuning of 13B models

Platform	Price ⇅	24 hours	Inventory	Billing Information
Vast.ai (USA)Global lowest priceNorth America Stability 100·A	¥1.53/hourCurrently Monitored Low PriceReal-timeSpot	¥37/day	sufficient	Vast.ai real-time spot price converted (RTX 5090, USD × 7.2); subject to change at any time	Rent Now
Salad (USA)Bulk lowest pricesNorth America Stability 74·C	¥1.80/hourReal-timeSpot	¥43/day	sufficient	Salad Distributed Cloud lowest priority tier: $0.250/hr (32GB VRAM; 4 tiers total—higher tiers offer greater stability)	Rent Now
SuanjiayunCurrent Low PriceChina Stability 100·A	¥2.68/hourReal-timePay-as-you-go	¥64/day	sufficient	Suanjiayun official website lists ¥2.68/hour (original ¥3.28; additional discounts available for long-term rentals)	Rent Now
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥2.93/hourReal-timePay-as-you-go	¥70/day	sufficient	AutoDL pay-as-you-go lowest price ¥2.93/hour (official marketplace RTX 5090, 43 idle GPUs)	Rent Now
MatPoolListed-price competitivenessChina Stability 100·A	¥2.99/hourReal-timePay-as-you-go	¥72/day	sufficient	MatPool pay-as-you-go pricing: ¥2.99/hour (Zone 17, clearly marked on official site)	Rent Now
AIStationChina Data collection in progress: 5/30 days	¥3.16/hourReal-timePay-as-you-go	¥76/day	sufficient	AIStation AI Galaxy (China) Low-Price Zone listed pay-as-you-go price: ¥3.16/hour (RTX 5090 32G)	Rent Now
CompshareCash RebateChina Stability 100·A	¥3.36/hourReal-timePay-as-you-go	¥81/day	sufficient	Compshare pay-as-you-go lowest configuration ¥3.36/hour (lower prices available for daily/monthly/spot instances on official website; 5% discount for universities and enterprises)	Rent Now
ParateraRTX 3090 Low PriceChina Data Accumulation: 1/30 Days	¥4.50/hourReal-timePay-as-you-go	¥108/day	sufficient	Paratera pay-as-you-go real-time price ¥4.50/hour	Rent Now
Cloudrift (United States)Structured, transparent pricingNorth America Data accumulation in progress: 2/30 days	¥4.68/hourReal-timePay-as-you-go	¥112/day	sufficient	Cloudrift official website JSON-LD RTX 5090 On-Demand Hourly Rate converted from $0.65/hr	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥4.97/hourReal-timeSpot	¥119/day	sufficient	RunPod Community Cloud lowest price: $0.69/hr (NVIDIA GeForce RTX 5090; Secure Cloud: $1.19/hr)	Rent Now
TensorDock (USA)Global Low PricesNorth America Manually maintained	¥4.10/hourReference price · Verification pendingPay-as-you-go	¥98/day	Tight	Market price ≈ $0.57 × 7.2; inventory fluctuates significantly ⚠ Note on pricing methodology: this site uses the official “minimum 8-GPU rental” pricing tier; these figures are not fully comparable to single-GPU pay-as-you-go rates from other platforms—provided for reference only.	Rent Now

A100 80GB80GB · Data center

Full-parameter fine-tuning for 7B models; quantized inference for 70B models; medium-scale training tasks

Platform	Price ⇅	24 hours	Inventory	Billing Information
Paperspace (United States)Fastest to get startedNorth America Data collection in progress: 12/30 days	¥5.47/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥131/day	sufficient	Paperspace (United States, a DigitalOcean company) on-demand: $0.76/hour (A100 80GB, hourly billing)	Rent Now
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥6.41/hourReal-timeSpot	¥154/day	sufficient	Jarvislabs (India) Spot: $0.89/GPU·hr (NVIDIA A100 80GB, billed per minute)	Rent Now
GpushareStable and EstablishedChina Stability 100·A	¥6.66/hourReal-timePay-as-you-go	¥160/day	sufficient	Gpushare pay-as-you-go real-time price: ¥6.66/hour (original price ¥15.00)	Rent Now
Thunder Compute (USA)Per-minute billingNorth America Stability 71·C	¥7.85/hourReal-timePay-as-you-go	¥188/day	sufficient	Thunder Compute (USA) pay-as-you-go at $1.09/hr (A100 80GB, customizable storage)	Rent Now
Verda (Finland)European value-for-moneyEurope Stability 74·C	¥8.09/hourReal-timeSpot	¥194/day	sufficient	Verda (Finland, formerly DataCrunch) Spot: $1.124/hr (equivalent), On-Demand: $2.248/hr (1× A100 SXM4 80GB)	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥8.57/hourReal-timeSpot	¥206/day	sufficient	RunPod Community Cloud lowest price: $1.19/hr (NVIDIA A100 80GB PCIe; Secure Cloud: $1.79/hr)	Rent Now
Massed Compute (United States)Transparent pricing for high-end GPUsNorth America Data accumulation in progress: 2/30 days	¥9.72/hourReal-timePay-as-you-go	¥233/day	sufficient	Massed Compute official website: A100 (80GB) full node at $1.35/hr ÷ 1 GPU = $1.35/GPU/hr (converted)	Rent Now
SuanjiayunCurrent Low PriceChina Stability 100·A	¥9.80/hourReal-timePay-as-you-go	¥235/day	sufficient	Suanjiayun official site lists ¥9.80/hour (original ¥12.00; additional discounts available for long-term rentals).	Rent Now
Gcore (Luxembourg)European edge cloudEurope Stability 71·C	¥10.61/hourReal-timePay-as-you-go	¥255/day	sufficient	Gcore (Luxembourg, global edge cloud): pay-as-you-go rate starting at €1.36/hour (A100)	Rent Now
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥10.73/hour▲67.4%Real-timePay-as-you-go	¥258/day	sufficient	Jarvislabs (India) On-Demand: $1.49/GPU·hr (NVIDIA A100 80GB, billed per minute, billing pauses when suspended)	Rent Now
CompshareCash RebateChina Stability 100·A	¥10.75/hourReal-timePay-as-you-go	¥258/day	sufficient	Compshare’s lowest pay-as-you-go configuration: ¥10.75/hour (even lower rates available for daily/monthly/spot plans on official site; 5% discount for universities and enterprises).	Rent Now
Hyperstack (UK)Frequent Entrant on the Ranking ListEurope Stability 74·C	¥11.52/hourReal-timePay-as-you-go	¥276/day	sufficient	Hyperstack (UK) on-demand: $1.60/hour (NVIDIA A100 SXM, 80GB VRAM)	Rent Now
Crusoe (USA)Green energyNorth America Stability 71·C	¥14.40/hourReal-timePay-as-you-go	¥346/day	sufficient	Crusoe (USA, powered by clean energy): pay-as-you-go rate starting at $2.00/GPU·hr (NVIDIA A100 80GB)	Rent Now
ParateraRTX 3090 Low PriceChina Data Accumulation: 1/30 Days	¥16.16/hourReal-timePay-as-you-go	¥388/day	sufficient	Paratera pay-as-you-go real-time price ¥16.16/hour	Rent Now
Lambda (USA)Established Prebuilt SystemsNorth America Data collection in progress: 12/30 days	¥20.09/hourReal-timePay-as-you-go	¥482/day	sufficient	Lambda (United States) official on-demand price: $2.79 per GPU-hour (NVIDIA A100 SXM 80GB, billed per minute, no outbound bandwidth fee)	Rent Now
TensorDock (USA)Global Low PricesNorth America Manually maintained	¥10.80/hourReference price · Verification pendingPay-as-you-go	¥259/day	sufficient	PCIe $1.50×7.2；SXM $1.80 ⚠ Note on pricing methodology: this site uses the official “minimum 8-GPU rental” pricing tier; these figures are not fully comparable to single-GPU pay-as-you-go rates from other platforms—provided for reference only.	Rent Now

A800 80GB80GB · Data center

A100 Domestic Compliance Edition — Enterprise Training Workhorse

Platform	Price ⇅	24 hours	Inventory	Billing Information
MornAILong-Term Rental Offers Better ValueChina Stability 100·A	¥5.15/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥124/day	sufficient	MornAI official website daily rate: ¥123.53/day ÷24 (NVIDIA A800 80G NVLink); pay-as-you-go hourly rate may be slightly higher.	Rent Now
GpushareStable and EstablishedChina Stability 100·A	¥6.80/hourReal-timePay-as-you-go	¥163/day	sufficient	Gpushare pay-as-you-go real-time price: ¥6.80/hour (original price: ¥15.00)	Rent Now
CompshareCash RebateChina Stability 100·A	¥7.36/hourReal-timePay-as-you-go	¥177/day	sufficient	Compshare pay-as-you-go minimum configuration starts at ¥7.36/hour (lower rates for daily/monthly/spot instances available on official website; 5% discount for universities and enterprises)	Rent Now
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥8.40/hourReal-timePay-as-you-go	¥202/day	sufficient	AutoDL pay-as-you-go lowest price ¥8.40/hour (official marketplace A800-80GB-NVLink, idle units: 18)	Rent Now

H800 80GB80GB · Datacenter flagship

Large model training/inference: H100 domestic compliance version

Platform	Price ⇅	24 hours	Inventory	Billing Information
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥10.50/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥252/day	sufficient	AutoDL pay-as-you-go lowest price: ¥10.50/hour (official marketplace H800, 9 idle units)	Rent Now

H100 80GB80GB · Datacenter flagship

Overseas platforms’ flagship GPUs for large model training

Platform	Price ⇅	24 hours	Inventory	Billing Information
Paperspace (United States)Fastest to get startedNorth America Data collection in progress: 12/30 days	¥8.28/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥199/day	sufficient	Paperspace (United States, a DigitalOcean company) on-demand: $1.15/hour (H100 80GB, hourly billing)	Rent Now
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥11.09/hourReal-timeSpot	¥266/day	sufficient	Jarvislabs (India) Spot: $1.54/GPU·hr (NVIDIA H100 SXM, billed per minute)	Rent Now
PPIO Paiou CloudLow-cost domestic spot pricingChina Stability 71·C	¥11.88/hourReal-timeSpot	¥285/day	sufficient	PPIO Paiou Cloud GPU spot price: ¥11.88/hr (H100; default 1-hour protection period, 1-hour termination notice)	Rent Now
Latitude.sh (Canada)Bare-metal transparent pricingNorth America Data accumulation in progress: 2/30 days	¥12.10/hourReal-timePay-as-you-go	¥290/day	sufficient	Latitude.sh official pricing page: g3-h100-small starts at $1.68/hr (lowest-tier equivalent for same GPU model)	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥14.33/hourReal-timeSpot	¥344/day	sufficient	RunPod Community Cloud lowest price: $1.99/hr (NVIDIA H100 PCIe; Secure Cloud: $2.89/hr)	Rent Now
Vast.ai (USA)Global lowest priceNorth America Stability 100·A	¥16.82/hourReal-timeSpot	¥404/day	sufficient	Vast.ai Real-time Spot Pricing Converted (H100 NVL, USD×7.2), subject to change	Rent Now
Verda (Finland)European value-for-moneyEurope Stability 74·C	¥17.26/hourReal-timeSpot	¥414/day	sufficient	Verda (Finland, formerly DataCrunch) Spot: $2.397/hr (equivalent), On-Demand: $4.793/hr (1× H100 SXM5 80GB)	Rent Now
Hyperstack (UK)Frequent Entrant on the Ranking ListEurope Stability 74·C	¥18.00/hourReal-timePay-as-you-go	¥432/day	sufficient	Hyperstack (UK) on-demand: $2.50/hour (NVIDIA H100, 80GB VRAM)	Rent Now
Massed Compute (United States)Transparent pricing for high-end GPUsNorth America Data accumulation in progress: 2/30 days	¥19.66/hourReal-timePay-as-you-go	¥472/day	sufficient	Massed Compute official website: H100 (80GB) full node at $2.73/hr ÷ 1 GPU = $2.73/GPU/hr (converted)	Rent Now
Gcore (Luxembourg)European edge cloudEurope Stability 71·C	¥21.29/hourReal-timePay-as-you-go	¥511/day	sufficient	Gcore (Luxembourg, global edge cloud): pay-as-you-go rate starting at €2.73/hour (H100)	Rent Now
Thunder Compute (USA)Per-minute billingNorth America Stability 71·C	¥23.04/hourReal-timePay-as-you-go	¥553/day	sufficient	Thunder Compute (USA) — pay-as-you-go at $3.2/hr (H100 SXM, customizable storage)	Rent Now
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥25.13/hour▲126.6%Real-timePay-as-you-go	¥603/day	sufficient	Jarvislabs (India) On-Demand: $3.49/GPU·hr (NVIDIA H100 SXM, billed per minute, billing pauses when suspended)	Rent Now
Crusoe (USA)Green energyNorth America Stability 71·C	¥28.08/hourReal-timePay-as-you-go	¥674/day	sufficient	Crusoe (USA, powered by clean energy): pay-as-you-go rate starting at $3.90/GPU·hr (NVIDIA H100 80GB)	Rent Now
Lambda (USA)Established Prebuilt SystemsNorth America Data collection in progress: 12/30 days	¥28.73/hourReal-timePay-as-you-go	¥690/day	sufficient	Lambda (United States) official on-demand price: $3.99 per GPU-hour (NVIDIA H100 SXM 80GB, billed per minute, no outbound bandwidth fee)	Rent Now
Nebius (Netherlands)Spot PricingEurope Stability 73·C	¥32.40/hourReal-timePay-as-you-go	¥778/day	sufficient	Nebius (Netherlands) on-demand: $4.50/GPU·hr (NVIDIA HGX H100)	Rent Now
TensorDock (USA)Global Low PricesNorth America Manually maintained	¥16.20/hourReference price · Verification pendingPay-as-you-go	¥389/day	sufficient	SXM $2.25×7.2 — Industry’s lowest on-demand tier ⚠ Note on pricing methodology: this site uses the official “minimum 8-GPU rental” pricing tier; these figures are not fully comparable to single-GPU pay-as-you-go rates from other platforms—provided for reference only.	Rent Now

L40S 48GB48GB · Data center

48 GB VRAM sweet-spot GPU: ideal for 70B quantized inference and batch inference

Platform	Price ⇅	24 hours	Inventory	Billing Information
Cloudrift (United States)Structured, transparent pricingNorth America Data accumulation in progress: 2/30 days	¥4.54/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥109/day	sufficient	Cloudrift official website JSON-LD L40S On-Demand Hourly Rate converted from $0.63/hr	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥5.69/hourReal-timeSpot	¥137/day	sufficient	RunPod Community Cloud lowest price: $0.79/hr (NVIDIA L40S; Secure Cloud: $1.09/hr)	Rent Now
Verda (Finland)European value-for-moneyEurope Stability 74·C	¥6.84/hourReal-timeSpot	¥164/day	sufficient	Verda (Finland, formerly DataCrunch) Spot: $0.9505/hr (equivalent), On-Demand: $1.901/hr (1× L40S 48GB)	Rent Now
Massed Compute (United States)Transparent pricing for high-end GPUsNorth America Data accumulation in progress: 2/30 days	¥6.98/hourReal-timePay-as-you-go	¥168/day	sufficient	Massed Compute official website: L40S (48GB) full node at $0.97/hr ÷ 1 GPU = $0.97/GPU/hr (converted)	Rent Now
Gcore (Luxembourg)European edge cloudEurope Stability 71·C	¥9.91/hourReal-timePay-as-you-go	¥238/day	sufficient	Gcore (Luxembourg, global edge cloud): pay-as-you-go rate starting at €1.27/hour (L40S)	Rent Now
Crusoe (USA)Green energyNorth America Stability 71·C	¥10.80/hourReal-timePay-as-you-go	¥259/day	sufficient	Crusoe (USA, powered by clean energy): pay-as-you-go rate starting at $1.50/GPU·hr (NVIDIA L40S 48GB)	Rent Now
Nebius (Netherlands)Spot PricingEurope Stability 73·C	¥11.16/hourReal-timePay-as-you-go	¥268/day	sufficient	Nebius (Netherlands) on-demand pricing: $1.55/GPU·hr (NVIDIA L40S with Intel CPU) (Intel CPU instance)	Rent Now
Vultr (USA)Maximum number of data centersNorth America Data collection in progress: 12/30 days	¥12.03/hourReal-timePay-as-you-go	¥289/day	sufficient	Vultr official on-demand price: $1.671/GPU·hour (NVIDIA L40S, 48GB VRAM full server, prorated per GPU)	Rent Now
SuanjiayunCurrent Low PriceChina Stability 100·A	¥5.08/hourPay-as-you-go	¥122/day	sufficient	Suanjiayun’s official website lists ¥5.08/hour (original price ¥7.98); domestic outbound tests conducted on 2026-10-30 confirmed Singapore outbound is unreachable, so automatic 45-minute refresh is temporarily unavailable—this price is treated as a reference value and excluded from the lowest-price ranking.	Rent Now

T4 16GB16GB · Entry-level inference

Ideal for students & light inference: 7B quantized inference, slow SD image generation, course labs

Platform	Price ⇅	24 hours	Inventory	Billing Information
MatPoolListed-price competitivenessChina Stability 100·A	¥1.35/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥32/day	sufficient	MatPool pay-per-minute pricing: ¥1.35/hour (Zone 0, clearly listed on official website)	Rent Now
ParateraRTX 3090 Low PriceChina Data Accumulation: 1/30 Days	¥3.50/hourReal-timePay-as-you-go	¥84/day	sufficient	Paratera pay-as-you-go real-time price ¥3.50/hour	Rent Now

A6000 48GB48 GB · Professional GPU

Large 48 GB VRAM: 70B quantized inference, QLoRA fine-tuning of 70B models, rendering

Platform	Price ⇅	24 hours	Inventory	Billing Information
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥2.38/hourCurrently Monitored Low PriceReal-timeSpot	¥57/day	sufficient	RunPod Community Cloud lowest price: $0.33/hr (NVIDIA RTX A6000; Secure Cloud: $0.59/hr)	Rent Now
Thunder Compute (USA)Per-minute billingNorth America Stability 71·C	¥2.52/hourReal-timePay-as-you-go	¥60/day	sufficient	Thunder Compute (USA) pay-as-you-go at $0.35/hr (RTX A6000, customizable storage)	Rent Now
MatPoolListed-price competitivenessChina Stability 100·A	¥2.85/hourReal-timePay-as-you-go	¥68/day	sufficient	MatPool pay-per-minute pricing: ¥2.85/hour (Zone 0, clearly listed on official website)	Rent Now
Verda (Finland)European value-for-moneyEurope Stability 74·C	¥2.98/hourReal-timeSpot	¥72/day	sufficient	Verda (Finland, formerly DataCrunch) Spot: $0.4140/hr (equivalent), On-Demand: $0.8280/hr (1× RTX A6000 48GB)	Rent Now
GpushareStable and EstablishedChina Stability 100·A	¥3.06/hour▲39.1%Real-timePay-as-you-go	¥73/day	sufficient	Gpushare pay-as-you-go real-time price: $3.06/hour (original $6.00)	Rent Now
Massed Compute (United States)Transparent pricing for high-end GPUsNorth America Data accumulation in progress: 2/30 days	¥4.10/hourReal-timePay-as-you-go	¥98/day	sufficient	Massed Compute official website: RTX A6000 (48GB) full node at $0.57/hr ÷ 1 GPU = $0.57/GPU/hr (converted)	Rent Now
Lambda (USA)Established Prebuilt SystemsNorth America Data collection in progress: 12/30 days	¥7.85/hourReal-timePay-as-you-go	¥188/day	sufficient	Lambda (United States) official on-demand price: $1.09 per GPU-hour (NVIDIA A6000 48GB, billed per minute, no outbound bandwidth fee)	Rent Now
Paperspace (United States)Fastest to get startedNorth America Data collection in progress: 12/30 days	¥9.94/hourReal-timePay-as-you-go	¥239/day	sufficient	Paperspace (United States, a DigitalOcean company) on-demand: $1.38/hour (A6000 48GB, hourly billing)	Rent Now
TensorDock (USA)Global Low PricesNorth America Manually maintained	¥3.24/hourReference price · Verification pendingPay-as-you-go	¥78/day	sufficient	$0.45×7.2 ⚠ Note on pricing methodology: this site uses the official “minimum 8-GPU rental” pricing tier; these figures are not fully comparable to single-GPU pay-as-you-go rates from other platforms—provided for reference only.	Rent Now

H200 141GB141GB · Data Center Flagship

H100 upgraded version with 141GB HBM3e—ideal for full-parameter 70B training and ultra-long-context inference.

Platform	Price ⇅	24 hours	Inventory	Billing Information
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥16.49/hourCurrently Monitored Low PriceReal-timeSpot	¥396/day	sufficient	Jarvislabs (India) Spot: $2.29/GPU·hr (NVIDIA H200 SXM, billed per minute)	Rent Now
Verda (Finland)European value-for-moneyEurope Stability 74·C	¥22.50/hourReal-timeSpot	¥540/day	sufficient	Verda (Finland, formerly DataCrunch) Spot: $3.125/hr (equivalent), On-Demand: $6.250/hr (1× H200 SXM5 141GB)	Rent Now
Gcore (Luxembourg)European edge cloudEurope Stability 71·C	¥24.80/hourReal-timePay-as-you-go	¥595/day	sufficient	Gcore (Luxembourg, global edge cloud): pay-as-you-go rate starting at €3.18/hour (H200)	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥25.85/hourReal-timeSpot	¥620/day	sufficient	RunPod Community Cloud lowest price: $3.59/hr (NVIDIA H200; Secure Cloud: $5.29/hr)	Rent Now
Massed Compute (United States)Transparent pricing for high-end GPUsNorth America Data accumulation in progress: 2/30 days	¥26.06/hourReal-timePay-as-you-go	¥625/day	sufficient	Massed Compute official website: H200 NVL (141GB) full node at $3.62/hr ÷ 1 GPU = $3.62/GPU/hr (converted)	Rent Now
Hyperstack (UK)Frequent Entrant on the Ranking ListEurope Stability 74·C	¥28.73/hourReal-timePay-as-you-go	¥690/day	sufficient	Hyperstack (UK) on-demand: $3.99/hour (NVIDIA H200 SXM, 141GB VRAM)	Rent Now
Crusoe (USA)Green energyNorth America Stability 71·C	¥30.89/hourReal-timePay-as-you-go	¥741/day	sufficient	Crusoe (US, renewable energy) on-demand: $4.29/GPU·hr (NVIDIA H200 141GB)	Rent Now
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥33.05/hour▲100.4%Real-timePay-as-you-go	¥793/day	sufficient	Jarvislabs (India) On-Demand: $4.59/GPU·hr (NVIDIA H200 SXM, billed per minute, billing pauses when suspended)	Rent Now
Nebius (Netherlands)Spot PricingEurope Stability 73·C	¥38.88/hourReal-timePay-as-you-go	¥933/day	sufficient	Nebius (Netherlands) on-demand: $5.40/GPU·hr (NVIDIA HGX H200)	Rent Now

B200 180GB180GB · Data Center Flagship

Blackwell flagship with 180GB VRAM—the new leader for training models with hundreds of billions of parameters.

Platform	Price ⇅	24 hours	Inventory	Billing Information
Verda (Finland)European value-for-moneyEurope Stability 74·C	¥31.93/hourCurrently Monitored Low PriceReal-timeSpot	¥766/day	sufficient	Verda (Finland, formerly DataCrunch) Spot: $4.435/hr (equivalent), On-Demand: $8.869/hr (1× B200 SXM6 180GB)	Rent Now
Massed Compute (United States)Transparent pricing for high-end GPUsNorth America Data accumulation in progress: 2/30 days	¥39.11/hourReal-timePay-as-you-go	¥939/day	sufficient	Massed Compute official website: B200 SXM6 full node at $43.46/hr ÷ 8 GPUs = $5.43/GPU/hr (converted)	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥43.06/hourReal-timeSpot	¥1033/day	sufficient	RunPod Community Cloud lowest price: $5.98/hr (NVIDIA B200; Secure Cloud: $7.99/hr)	Rent Now
Hyperstack (UK)Frequent Entrant on the Ranking ListEurope Stability 74·C	¥43.20/hourReal-timePay-as-you-go	¥1037/day	sufficient	Hyperstack (UK) on-demand: $6.00/hour (NVIDIA B200, 192GB VRAM)	Rent Now
Lambda (USA)Established Prebuilt SystemsNorth America Data collection in progress: 12/30 days	¥48.17/hourReal-timePay-as-you-go	¥1156/day	sufficient	Lambda (United States) official on-demand price: $6.69 per GPU-hour (NVIDIA B200 SXM6 180GB, billed per minute, no outbound bandwidth fee)	Rent Now
Nebius (Netherlands)Spot PricingEurope Stability 73·C	¥61.20/hourReal-timePay-as-you-go	¥1469/day	sufficient	Nebius (Netherlands) on-demand: $8.50/GPU·hr (NVIDIA HGX B200)	Rent Now

L4 24GB24GB · Entry-level Inference

Low-power inference GPU; high cost-efficiency for video transcoding and lightweight inference

Platform	Price ⇅	24 hours	Inventory	Billing Information
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥2.09/hourCurrently Monitored Low PriceReal-timeSpot	¥50/day	sufficient	Jarvislabs (India) Spot: $0.29/GPU·hr (NVIDIA L4, billed per minute)	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥3.17/hourReal-timeSpot	¥76/day	sufficient	RunPod Community Cloud lowest price: $0.44/hr (NVIDIA L4; Secure Cloud: $0.59/hr)	Rent Now
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥3.17/hour▲51.7%Real-timePay-as-you-go	¥76/day	sufficient	Jarvislabs (India) On-Demand: $0.44/GPU·hr (NVIDIA L4, billed per minute, billing pauses when suspended)	Rent Now
ParateraRTX 3090 Low PriceChina Data Accumulation: 1/30 Days	¥4.10/hourReal-timePay-as-you-go	¥98/day	sufficient	Paratera pay-as-you-go real-time price ¥4.10/hour	Rent Now

L40 48GB48GB · Data center

48GB VRAM General-purpose GPU—Balances Rendering and Inference

Platform	Price ⇅	24 hours	Inventory	Billing Information
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥4.97/hourCurrently Monitored Low PriceReal-timeSpot	¥119/day	sufficient	RunPod Community Cloud lowest price: $0.69/hr (NVIDIA L40; Secure Cloud: $0.82/hr)	Rent Now
Thunder Compute (USA)Per-minute billingNorth America Stability 71·C	¥5.69/hourReal-timePay-as-you-go	¥137/day	sufficient	Thunder Compute (USA) pay-as-you-go at $0.79/hr (L40, customizable storage)	Rent Now
Massed Compute (United States)Transparent pricing for high-end GPUsNorth America Data accumulation in progress: 2/30 days	¥6.19/hourReal-timePay-as-you-go	¥149/day	sufficient	Massed Compute official website: L40 (48GB) full node at $0.86/hr ÷ 1 GPU = $0.86/GPU/hr (converted)	Rent Now

A40 48GB48GB · Data center

Ampere-generation 48GB GPU, widely available on the secondary market, low rental cost

Platform	Price ⇅	24 hours	Inventory	Billing Information
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥2.52/hourCurrently Monitored Low PriceReal-timeSpot	¥60/day	sufficient	RunPod Community Cloud lowest price: $0.35/hr (NVIDIA A40; Secure Cloud: $0.59/hr)	Rent Now

H20 96GB96GB · Data Center

Domestically compliant H100 variant with 96GB VRAM—one of China’s primary GPUs for local training.

Platform	Price ⇅	24 hours	Inventory	Billing Information
PPIO Paiou CloudLow-cost domestic spot pricingChina Stability 71·C	¥3.40/hourCurrently Monitored Low PriceReal-timeSpot	¥82/day	sufficient	PPIO Paiou Cloud GPU spot price: ¥3.40/hr (H20; default 1-hour protection period, 1-hour termination notice)	Rent Now
CompshareCash RebateChina Stability 100·A	¥7.50/hourReal-timePay-as-you-go	¥180/day	sufficient	Compshare pay-as-you-go minimum configuration: ¥7.50/hour (lower rates available for daily/monthly/spot instances—see official website; 5% discount for universities and enterprises)	Rent Now
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥10.50/hourReal-timePay-as-you-go	¥252/day	sufficient	AutoDL pay-as-you-go lowest price ¥10.50/hour (official marketplace H20-NVLink, idle units: 7)	Rent Now

RTX 4090D 24GB24GB · Consumer flagship

RTX 4090 Domestic Compliance Edition: ~90% Performance of RTX 4090, Typically Priced Lower

Platform	Price ⇅	24 hours	Inventory	Billing Information
SuanjiayunCurrent Low PriceChina Stability 100·A	¥1.14/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥27/day	sufficient	Suanjiayun official website lists ¥1.14/hour (original ¥1.98; additional discounts available for long-term rentals)	Rent Now
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥1.98/hourReal-timePay-as-you-go	¥48/day	sufficient	AutoDL pay-as-you-go lowest price ¥1.98/hour (official marketplace RTX 4090D, idle units: 49)	Rent Now

RTX PRO 6000 96GB96GB · Professional GPU flagship

Blackwell professional GPU with 96GB ultra-large VRAM—capable of running full-parameter 70B models on a single card.

Platform	Price ⇅	24 hours	Inventory	Billing Information
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥7.35/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥176/day	sufficient	AutoDL pay-as-you-go lowest price ¥7.35/hour (official marketplace RTX PRO 6000, idle units: 12)	Rent Now
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥8.28/hourReal-timeSpot	¥199/day	sufficient	Jarvislabs (India) Spot: $1.15/GPU·hr (NVIDIA RTX Pro 6000 Blackwell, billed per minute)	Rent Now
Verda (Finland)European value-for-moneyEurope Stability 74·C	¥9.53/hourReal-timeSpot	¥229/day	sufficient	Verda (Finland, formerly DataCrunch) Spot: $1.323/hr (equivalent), On-Demand: $2.645/hr (1× RTX PRO 6000 96GB)	Rent Now
Nebius (Netherlands)Spot PricingEurope Stability 73·C	¥12.96/hourReal-timePay-as-you-go	¥311/day	sufficient	Nebius (Netherlands) on-demand: $1.80/GPU·hr (NVIDIA RTX PRO 6000)	Rent Now
Hyperstack (UK)Frequent Entrant on the Ranking ListEurope Stability 74·C	¥13.32/hourReal-timePay-as-you-go	¥320/day	sufficient	Hyperstack (UK) on-demand: $1.85/hour (NVIDIA RTX Pro 6000 SE, 96GB VRAM)	Rent Now
Massed Compute (United States)Transparent pricing for high-end GPUsNorth America Data accumulation in progress: 2/30 days	¥15.77/hourReal-timePay-as-you-go	¥378/day	sufficient	Massed Compute official website: RTX PRO 6000 Blackwell (96GB) full node at $2.19/hr ÷ 1 GPU = $2.19/GPU/hr (converted)	Rent Now
Jarvislabs (India)South Asia PreferredAsia-Pacific Data collection in progress: 5/30 days	¥15.77/hour▲90.5%Real-timePay-as-you-go	¥378/day	sufficient	Jarvislabs (India) On-Demand: $2.19/GPU·hr (NVIDIA RTX Pro 6000 Blackwell, billed per minute, billing pauses when suspended)	Rent Now

RTX 6000 Ada 48GB48 GB · Professional GPU

Ada-generation professional flagship GPU with 48GB VRAM, versatile for rendering, inference, and fine-tuning

Platform	Price ⇅	24 hours	Inventory	Billing Information
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥5.33/hourCurrently Monitored Low PriceReal-timeSpot	¥128/day	sufficient	RunPod Community Cloud lowest price: $0.74/hr (NVIDIA RTX 6000 Ada Generation; Secure Cloud: $0.99/hr)	Rent Now
Verda (Finland)European value-for-moneyEurope Stability 74·C	¥5.39/hourReal-timeSpot	¥129/day	sufficient	Verda (Finland, formerly DataCrunch) Spot: $0.7490/hr (equivalent), On-Demand: $1.498/hr (1× RTX 6000 Ada 48GB)	Rent Now
Massed Compute (United States)Transparent pricing for high-end GPUsNorth America Data accumulation in progress: 2/30 days	¥5.69/hourReal-timePay-as-you-go	¥137/day	sufficient	Massed Compute official website: RTX 6000 ADA (48GB) full node at $0.79/hr ÷ 1 GPU = $0.79/GPU/hr (converted)	Rent Now

V100 16GB16GB · Entry-level Training

Legacy training GPU; extremely low rental cost—sufficient for coursework labs and legacy framework reproduction

Platform	Price ⇅	24 hours	Inventory	Billing Information
Verda (Finland)European value-for-moneyEurope Stability 74·C	¥0.85/hourCurrently Monitored Low PriceReal-timeSpot	¥20/day	sufficient	Verda (Finland, formerly DataCrunch): spot $0.1180/hr equivalent, on-demand $0.2360/hr (1× Tesla V100 16GB)	Rent Now
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥1.37/hourReal-timeSpot	¥33/day	sufficient	RunPod Community Cloud lowest price: $0.19/hr (Tesla V100-PCIE-16GB; Secure Cloud: $0.00/hr)	Rent Now
Cloudrift (United States)Structured, transparent pricingNorth America Data accumulation in progress: 2/30 days	¥1.80/hourReal-timePay-as-you-go	¥43/day	sufficient	Cloudrift official website JSON-LD V100 SXM2 On-Demand Hourly Rate converted from $0.25/hr	Rent Now
Lambda (USA)Established Prebuilt SystemsNorth America Data collection in progress: 12/30 days	¥5.69/hourReal-timePay-as-you-go	¥137/day	sufficient	Lambda (United States) official on-demand price: $0.79 per GPU-hour (NVIDIA Tesla V100 16GB, billed per minute, no outbound bandwidth fee)	Rent Now

MI300X 192GB192GB · Data Center Flagship

AMD flagship GPU with 192GB VRAM—the industry’s largest—ROCm ecosystem, large-model inference

Platform	Price ⇅	24 hours	Inventory	Billing Information
RunPod (USA)Top overseas favoriteNorth America Stability 73·C	¥3.60/hourCurrently Monitored Low PriceReal-timeSpot	¥86/day	sufficient	RunPod Community Cloud lowest price: $0.50/hr (AMD Instinct MI300X OAM; Secure Cloud: $2.39/hr)	Rent Now
Crusoe (USA)Green energyNorth America Stability 71·C	¥24.84/hourReal-timePay-as-you-go	¥596/day	sufficient	Crusoe (US, renewable energy) on-demand: $3.45/GPU·hr (AMD MI300X 192GB)	Rent Now

RTX 6000 24GB24 GB · Professional GPU

Professional card based on Turing architecture — 24GB VRAM, passively cooled; suitable for workstations and lightweight inference, but lower cost-effectiveness than RTX 4090

Platform	Price ⇅	24 hours	Inventory	Billing Information
Akamai / Linode (United States)Stable, large-scale providersNorth America Data collection in progress: 12/30 days	¥10.80/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥259/day	sufficient	Akamai (formerly Linode) listed on-demand price: $1.50 per GPU·hour (Dedicated 32GB + RTX6000 GPU x1; single-GPU configuration priced per GPU)	Rent Now

RTX 4000 Ada 20GB20 GB · Entry-level professional GPU

Ada architecture entry-level professional GPU: 20 GB VRAM, low power consumption—ideal for small-model inference, teaching, and development debugging

Platform	Price ⇅	24 hours	Inventory	Billing Information
Akamai / Linode (United States)Stable, large-scale providersNorth America Data collection in progress: 12/30 days	¥3.74/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥90/day	sufficient	Akamai (formerly Linode) listed on-demand price: $0.52 per GPU·hour (RTX4000 Ada x1 Small; single-GPU configuration priced per GPU)	Rent Now

V100 32GB32GB · Entry-level training

V100 with 32GB VRAM enables larger batch sizes and smoother training; sufficient for legacy framework reproduction and course labs

Platform	Price ⇅	24 hours	Inventory	Billing Information
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥1.98/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥48/day	sufficient	AutoDL pay-as-you-go lowest price ¥1.98/hour (official marketplace V100-32GB, idle units: 53)	Rent Now
ParateraRTX 3090 Low PriceChina Data Accumulation: 1/30 Days	¥3.10/hourReal-timePay-as-you-go	¥74/day	sufficient	Paratera pay-as-you-go real-time price ¥3.10/hour (normalized from full 8-GPU node price ¥24.84 ÷ 8)	Rent Now

A10 24GB24GB · Entry-level Inference

Professional GPU for graphics and inference—24GB VRAM, balanced power consumption and cost.

Platform	Price ⇅	24 hours	Inventory	Billing Information
ParateraRTX 3090 Low PriceChina Data Accumulation: 1/30 Days	¥6.12/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥147/day	sufficient	Paratera pay-as-you-go real-time price ¥6.12/hour	Rent Now

L20 48GB48GB · Data center

48GB high-VRAM inference GPU, ideal for large-model inference and long-context tasks

Platform	Price ⇅	24 hours	Inventory	Billing Information
ParateraRTX 3090 Low PriceChina Data Accumulation: 1/30 Days	¥11.25/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥270/day	sufficient	Paratera pay-as-you-go real-time price ¥11.25/hour	Rent Now

MTT S4000 48GB48GB · Domestic GPU

Moore Threads MTT S4000 (48GB, MUSA ecosystem); hourly rental of domestic GPUs, ideal for domestic substitution and IT innovation scenarios

Platform	Price ⇅	24 hours	Inventory	Billing Information
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥1.96/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥47/day	sufficient	AutoDL pay-as-you-go lowest price ¥1.96/hour (official marketplace MTT S4000, idle units: 31)	Rent Now

A100 40GB40GB · Data Center

A100 PCIe 40GB, the previous-generation workhorse GPU for data centers; the 40GB variant is typically priced lower than the 80GB version

Platform	Price ⇅	24 hours	Inventory	Billing Information
AutoDLMost Comprehensive EcosystemChina Stability 98·A	¥3.45/hourCurrently Monitored Low PriceReal-timePay-as-you-go	¥83/day	sufficient	AutoDL pay-as-you-go lowest price ¥3.45/hour (official marketplace A100-PCIE-40GB, idle units: 15)	Rent Now

Guides and Daily Digest GPU Price Index → · View All →

Daily Brief

GPU Rental, See at a Glance Which Provider Is Cheaper

Currently among the lowest prices network-wide

28 GPUs, lowest prices across all platforms displayed side-by-side

You sleep, it monitors the market

Full-network price scraping every 45 minutes

Sorting is always by price

We’ll notify you when the price drops to your target

Don’t own a single GPUOnly then do we show you real prices

Drop to your target price,You’ll be the first to know

RTX 309024GB · Consumer flagship

RTX 409024GB · Consumer flagship

RTX 509032GB · Next-gen consumer flagship

A100 80GB80GB · Data center

A800 80GB80GB · Data center

H800 80GB80GB · Datacenter flagship

H100 80GB80GB · Datacenter flagship

L40S 48GB48GB · Data center

T4 16GB16GB · Entry-level inference

A6000 48GB48 GB · Professional GPU

H200 141GB141GB · Data Center Flagship

B200 180GB180GB · Data Center Flagship

L4 24GB24GB · Entry-level Inference

L40 48GB48GB · Data center

A40 48GB48GB · Data center

H20 96GB96GB · Data Center

RTX 4090D 24GB24GB · Consumer flagship

RTX PRO 6000 96GB96GB · Professional GPU flagship

RTX 6000 Ada 48GB48 GB · Professional GPU

V100 16GB16GB · Entry-level Training

MI300X 192GB192GB · Data Center Flagship

RTX 6000 24GB24 GB · Professional GPU

RTX 4000 Ada 20GB20 GB · Entry-level professional GPU

V100 32GB32GB · Entry-level training

A10 24GB24GB · Entry-level Inference

L20 48GB48GB · Data center

MTT S4000 48GB48GB · Domestic GPU

A100 40GB40GB · Data Center

Guides and Daily Digest GPU Price Index → · View All →

GPU Price Comparison Bulletin, October 31: RTX 4090 as low as ¥0.87/hour

GPU Price Comparison Bulletin, Oct 30: RTX 4090 as low as ¥0.87/hour

GPU Price Comparison Bulletin, Oct 29: RTX 4090 as low as ¥0.87/hour

GPU Price Comparison Bulletin, October 28: RTX 4090 as low as ¥1.15/hour

Don’t own a single GPU
Only then do we show you real prices

Drop to your target price,
You’ll be the first to know