Simple, Transparent Pricing
Pay only for what you use. No hidden fees. No egress charges. Cancel anytime.
Up to 60% cheaper
on the same GPU models
Multi-Region Infrastructure
GPUs in APAC and beyond. Low-latency for developers in India, SEA, LatAm, and Europe.
60-second deploy
Browser-based. No SSH, no setup.
| GPU Model | Stock | CloudGPU | Market rate |
|---|---|---|---|
| H20 96GChina-spec Hopper, huge HBM3 | Low stock | $1.29/hr | — |
| Ascend 910B 64GNPU — CANN / MindSpore / PyTorch-NPU only | Low stock | $0.95/hr | — |
| A800 80GChina-spec A100 | Low stock | $1.09/hr | $2.17/hr |
| H800China-spec H100, 400Gb/s NVLink capped | Low stock | $1.79/hr | $3.99/hr |
Market rates observed April 2026 for the equivalent SKU. "—" = not offered on the open market.
Performance & Workload Fit
Price alone doesn't tell you which GPU is right. L20, H20, and Ascend 910B are China-spec cards that US clouds don't sell — they have unique VRAM / bandwidth profiles worth comparing.
| GPU | Arch | VRAM | FP16 TFLOPS | Mem BW | Best For |
|---|---|---|---|---|---|
H20 96G China-spec Hopper, huge HBM3 | Hopper | 96 GB | 148 | 4000 GB/s | 70B-120B inference, long-context, MoE models |
Ascend 910B 64G NPU — CANN / MindSpore / PyTorch-NPU only | Ascend | 64 GB | 320 | 1200 GB/s | Domestic-compliant training, MindSpore/PyTorch-NPU workloads |
A800 80G China-spec A100 | Ampere | 80 GB | 312 | 2039 GB/s | FP32/FP16 training 30B-70B, large-memory research |
H800 China-spec H100, 400Gb/s NVLink capped | Hopper | 80 GB | 989 | 3350 GB/s | Frontier training, 70B+ fine-tuning, FP8 inference |
VRAM determines model size
24 GB fits 7B at FP16 / 30B quantized. 48 GB fits 13B at FP16 / 70B quantized. 96 GB comfortably runs 70B FP16.
Mem bandwidth = tokens/sec
For LLM inference, throughput scales with memory bandwidth, not TFLOPS. H20's 4 TB/s beats A100 for serving large models.
FP16 TFLOPS = training speed
A100's 312 TFLOPS shines for fine-tuning. 910B's 320 is on paper — actual speed depends on CANN/MindSpore ecosystem maturity.
Cost Calculator
Pricing Plans
On-Demand
Pay as you go. No commitment.
Market Rate
- check Per-hour billing
- check Cancel anytime
- check No minimum spend
- check All GPU models
Prepaid Credits
Buy credits in bulk, get bonus.
$100 → $110
10% bonus credits
- check 10% bonus on top-ups of $100+
- check First top-up of $10 or more: $5 bonus
- check Credits never expire
- check Use on any GPU model
- check Works on GPUs and the model API
Enterprise
Steady volume, its own quote.
Talk to us
Priced on your volume
- check Reserved capacity
- check Volume rates
- check Multi-card clusters
- check A named contact
No hidden fees · No egress charges · Per-hour billing · Cancel anytime
How You Pay
Add credit to your account, then spend it on GPU rentals or the AI Gateway. One balance covers both. GPU time is billed by the minute; API calls by the token.
Debit or credit card
Coming soonCard payments are temporarily unavailable. Top up with USDT (TRC-20); business customers can pay by bank transfer.
When card payments go live the option appears on the Wallet page automatically. Nothing to set up on your side.
Bank Transfer
USD, EUR, GBP and 8 more currencies. From $50.
We give you our account details and an amount with a unique cent value, which is how your transfer is matched to your account. Usually 1–2 business days.
USDT (TRC-20)
Tether on the Tron network, credited 1:1.
Each account gets its own deposit address. Send from any wallet or exchange; the balance updates once the transfer is confirmed on-chain.
All prices are in US dollars. No setup fee, no egress charge, no minimum spend.