Console
/
GPU Market
Templates
Pricing API
English
▾
简体中文
English
日本語
한국어
Tiếng Việt
ไทย
Bahasa Indonesia
Bahasa Melayu
Filipino
Log in
Create account →
Templates
NVIDIA CUDA
PyTorch
TensorFlow CUDA
PyTorch NGC
NVIDIA RAPIDS
All-in-One App Studio
vLLM
vLLM Omni
SGLang
Llama.cpp
Ollama
Open WebUI (Ollama)
Oobabooga
HuggingFace TGI
Langflow (Ollama)
Axolotl
Unsloth Studio
Kohya's GUI
Flux Gym
Ostris AI Toolkit
ComfyUI
InvokeAI
SD WebUI Forge
SD WebUI A1111
Fooocus
SwarmUI
Wan2GP
Open-Sora
Whisper WebUI & API
Voicebox TTS
ACE Step 1.5
Linux Desktop
Pinokio
Ubuntu 22.04 VM
Ubuntu Desktop VM
Unreal Pixel Streaming
Hashcat CUDA
Browse 37 templates
Disk
64 GB · +$5.12/mo
On-demand
Interruptible −50%
Reserved −35%
Best value (performance/price)
Price: low to high
Price: high to low
Performance score
VRAM per GPU
Network speed
Reliability
Disk size
35
machines · 122 GPUs
(demo data)
Billing mode: On-demand · Billed per second
1× RTX 4080 SUPER
16 GB GDDR6X / GPU · AMD EPYC 7282 · PCIe 4.0 · CUDA 12.4 ·
m-2b6f4d52
PL
Warsaw ·
eu-central-2
Score
165
FP16 TFLOPS
209
vCPU · RAM
12 · 48G
NVMe disk
2,000G
3,200 MB/s
Network ↓ ↑
10 Gbps · 5 Gbps
Open ports
32
Reliability
98.46%
$0.203
/hr
≈ $153/mo (incl. 64 GB disk)
Market price
$0.290
/hr
Deploy now
Billed per second
1× RTX 4080 SUPER
16 GB GDDR6X / GPU · AMD EPYC 7282 · PCIe 4.0 · CUDA 13.0 ·
m-522da122
KR
Seoul ·
ap-northeast-2
Score
167
FP16 TFLOPS
209
vCPU · RAM
24 · 48G
NVMe disk
500G
2,800 MB/s
Network ↓ ↑
10 Gbps · 1 Gbps
Open ports
64
Reliability
99.90%
$0.203
/hr
≈ $153/mo (incl. 64 GB disk)
Market price
$0.290
/hr
Deploy now
Billed per second
1× RTX 4080 SUPER
16 GB GDDR6X / GPU · AMD EPYC 7282 · PCIe 4.0 · CUDA 12.8 ·
m-78a32c2d
DE
Frankfurt ·
eu-central-1
Score
165
FP16 TFLOPS
209
vCPU · RAM
8 · 128G
NVMe disk
2,000G
3,200 MB/s
Network ↓ ↑
10 Gbps · 500 Mbps
Open ports
32
Reliability
98.64%
$0.203
/hr
≈ $153/mo (incl. 64 GB disk)
Market price
$0.290
/hr
Deploy now
Billed per second
1× RTX 4090
24 GB GDDR6X / GPU · AMD Ryzen 9 7950X · PCIe 4.0 · CUDA 13.0 ·
m-2b404278
SE
Stockholm ·
eu-north-1
Score
263
FP16 TFLOPS
330
vCPU · RAM
24 · 48G
NVMe disk
8,000G
6,800 MB/s
Network ↓ ↑
5 Gbps · 500 Mbps
Open ports
64
Reliability
99.44%
$0.350
/hr
≈ $261/mo (incl. 64 GB disk)
Market price
$0.500
/hr
Deploy now
Billed per second
4× RTX 4090
24 GB GDDR6X / GPU · AMD Ryzen 9 7950X · PCIe 4.0 · CUDA 12.8 ·
m-15153c71
HK
Hong Kong ·
ap-east-1
Score
1,037
FP16 TFLOPS
1,320
vCPU · RAM
96 · 256G
NVMe disk
4,000G
2,800 MB/s
Network ↓ ↑
10 Gbps · 2.5 Gbps
Open ports
32
Reliability
98.22%
$1.400
/hr
≈ $1,027/mo (incl. 64 GB disk)
Market price
$2.000
/hr
Deploy now
Billed per second
4× RTX 4090
24 GB GDDR6X / GPU · AMD Ryzen 9 7950X · PCIe 4.0 · CUDA 12.8 ·
m-1467307e
KR
Seoul ·
ap-northeast-2
Score
1,044
FP16 TFLOPS
1,320
vCPU · RAM
64 · 192G
NVMe disk
1,000G
2,800 MB/s
Network ↓ ↑
2.5 Gbps · 5 Gbps
Open ports
16
Reliability
98.90%
$1.400
/hr
≈ $1,027/mo (incl. 64 GB disk)
Market price
$2.000
/hr
Deploy now
Billed per second
2× RTX 3090
24 GB GDDR6X / GPU · AMD Ryzen 9 5950X · PCIe 4.0 · CUDA 12.4 ·
m-685712d9
DE
Frankfurt ·
eu-central-1
Score
224
FP16 TFLOPS
284
vCPU · RAM
32 · 64G
NVMe disk
500G
6,800 MB/s
Network ↓ ↑
10 Gbps · 500 Mbps
Open ports
16
Reliability
98.63%
$0.336
/hr
≈ $250/mo (incl. 64 GB disk)
Market price
$0.480
/hr
Deploy now
Billed per second
1× RTX 3090
24 GB GDDR6X / GPU · Intel Core i9-13900K · PCIe 4.0 · CUDA 12.4 ·
m-46de9e39
HK
Hong Kong ·
ap-east-1
Score
112
FP16 TFLOPS
142
vCPU · RAM
12 · 32G
NVMe disk
8,000G
5,000 MB/s
Network ↓ ↑
5 Gbps · 500 Mbps
Open ports
100
Reliability
98.81%
$0.168
/hr
≈ $128/mo (incl. 64 GB disk)
Market price
$0.240
/hr
Deploy now
Billed per second
1× RTX 5090
32 GB GDDR7 / GPU · AMD EPYC 7302 · PCIe 5.0 · CUDA 12.4 ·
m-71e20a41
US
Dallas ·
us-central-1
Score
331
FP16 TFLOPS
419
vCPU · RAM
24 · 48G
NVMe disk
8,000G
2,800 MB/s
Network ↓ ↑
5 Gbps · 2.5 Gbps
Open ports
16
Reliability
98.64%
$0.504
/hr
≈ $373/mo (incl. 64 GB disk)
Market price
$0.720
/hr
Deploy now
Billed per second
8× RTX 5090
32 GB GDDR7 / GPU · AMD EPYC 7302 · PCIe 5.0 · CUDA 12.8 ·
m-1a754132
US
Dallas ·
us-central-1
Score
2,676
FP16 TFLOPS
3,352
vCPU · RAM
128 · 384G
NVMe disk
2,000G
6,800 MB/s
Network ↓ ↑
10 Gbps · 5 Gbps
Open ports
32
Reliability
99.78%
$4.032
/hr
≈ $2,948/mo (incl. 64 GB disk)
Market price
$5.760
/hr
Deploy now
Billed per second
8× RTX 5090
32 GB GDDR7 / GPU · AMD Threadripper 7970X · PCIe 5.0 · CUDA 12.4 ·
m-65195d2b
US
Los Angeles ·
us-west-1
Score
2,656
FP16 TFLOPS
3,352
vCPU · RAM
192 · 512G
NVMe disk
1,000G
2,800 MB/s
Network ↓ ↑
1 Gbps · 5 Gbps
Open ports
32
Reliability
99.05%
$4.032
/hr
≈ $2,948/mo (incl. 64 GB disk)
Market price
$5.760
/hr
Deploy now
Billed per second
4× H100 SXM
80 GB HBM3 / GPU · Intel Xeon Platinum 8480+ · PCIe 5.0 · CUDA 12.8 · NVLink ·
m-5736e023
DE
Frankfurt ·
eu-central-1
Score
3,124
FP16 TFLOPS
3,956
vCPU · RAM
64 · 128G
NVMe disk
500G
5,000 MB/s
Network ↓ ↑
1 Gbps · 1 Gbps
Open ports
100
Reliability
98.71%
$6.048
/hr
≈ $4,420/mo (incl. 64 GB disk)
Market price
$8.640
/hr
Deploy now
Billed per second
2× H100 SXM
80 GB HBM3 / GPU · AMD EPYC 9654 · PCIe 5.0 · CUDA 12.8 ·
m-75c9db6a
US
Dallas ·
us-central-1
Score
1,559
FP16 TFLOPS
1,978
vCPU · RAM
24 · 96G
NVMe disk
1,000G
6,800 MB/s
Network ↓ ↑
1 Gbps · 5 Gbps
Open ports
64
Reliability
98.55%
$3.024
/hr
≈ $2,213/mo (incl. 64 GB disk)
Market price
$4.320
/hr
Deploy now
Billed per second
4× H100 SXM
80 GB HBM3 / GPU · AMD EPYC 9654 · PCIe 5.0 · CUDA 13.0 · NVLink ·
m-5d545af2
BR
São Paulo ·
sa-east-1
Score
3,152
FP16 TFLOPS
3,956
vCPU · RAM
64 · 512G
NVMe disk
500G
5,000 MB/s
Network ↓ ↑
10 Gbps · 5 Gbps
Open ports
32
Reliability
99.58%
$6.048
/hr
≈ $4,420/mo (incl. 64 GB disk)
Market price
$8.640
/hr
Deploy now
Billed per second
1× H100 SXM
80 GB HBM3 / GPU · AMD EPYC 9554 · PCIe 5.0 · CUDA 12.8 ·
m-555581c1
KR
Seoul ·
ap-northeast-2
Score
790
FP16 TFLOPS
989
vCPU · RAM
8 · 64G
NVMe disk
8,000G
6,800 MB/s
Network ↓ ↑
1 Gbps · 2.5 Gbps
Open ports
32
Reliability
99.89%
$1.512
/hr
≈ $1,109/mo (incl. 64 GB disk)
Market price
$2.160
/hr
Deploy now
Billed per second
1× RTX 6000 Ada
48 GB GDDR6 / GPU · Intel Xeon Gold 6348 · PCIe 4.0 · CUDA 13.0 ·
m-222c6d98
DE
Frankfurt ·
eu-central-1
Score
287
FP16 TFLOPS
364
vCPU · RAM
8 · 32G
NVMe disk
1,000G
2,800 MB/s
Network ↓ ↑
5 Gbps · 5 Gbps
Open ports
16
Reliability
98.45%
$0.574
/hr
≈ $424/mo (incl. 64 GB disk)
Market price
$0.820
/hr
Deploy now
Billed per second
2× RTX 6000 Ada
48 GB GDDR6 / GPU · Intel Xeon Gold 6348 · PCIe 4.0 · CUDA 12.4 ·
m-172a37a7
US
Dallas ·
us-central-1
Score
579
FP16 TFLOPS
728
vCPU · RAM
24 · 256G
NVMe disk
2,000G
3,200 MB/s
Network ↓ ↑
2.5 Gbps · 5 Gbps
Open ports
100
Reliability
99.39%
$1.148
/hr
≈ $843/mo (incl. 64 GB disk)
Market price
$1.640
/hr
Deploy now
Billed per second
2× RTX 6000 Ada
48 GB GDDR6 / GPU · Intel Xeon Gold 6348 · PCIe 4.0 · CUDA 13.0 ·
m-5842dac3
US
Los Angeles ·
us-west-1
Score
580
FP16 TFLOPS
728
vCPU · RAM
16 · 96G
NVMe disk
8,000G
6,800 MB/s
Network ↓ ↑
1 Gbps · 2.5 Gbps
Open ports
32
Reliability
99.59%
$1.148
/hr
≈ $843/mo (incl. 64 GB disk)
Market price
$1.640
/hr
Deploy now
Billed per second
1× L40S
48 GB GDDR6 / GPU · AMD EPYC 7543 · PCIe 4.0 · CUDA 13.0 ·
m-1ab6994a
US
Dallas ·
us-central-1
Score
289
FP16 TFLOPS
362
vCPU · RAM
24 · 64G
NVMe disk
1,000G
2,800 MB/s
Network ↓ ↑
10 Gbps · 2.5 Gbps
Open ports
32
Reliability
99.80%
$0.616
/hr
≈ $455/mo (incl. 64 GB disk)
Market price
$0.880
/hr
Deploy now
Billed per second
2× L40S
48 GB GDDR6 / GPU · Intel Xeon Gold 6348 · PCIe 4.0 · CUDA 12.4 ·
m-6e7c9748
AU
Sydney ·
ap-southeast-2
Score
572
FP16 TFLOPS
724
vCPU · RAM
32 · 96G
NVMe disk
8,000G
6,800 MB/s
Network ↓ ↑
2.5 Gbps · 500 Mbps
Open ports
16
Reliability
98.68%
$1.232
/hr
≈ $904/mo (incl. 64 GB disk)
Market price
$1.760
/hr
Deploy now
Billed per second
8× L40S
48 GB GDDR6 / GPU · AMD EPYC 7763 · PCIe 4.0 · CUDA 13.0 ·
m-209f0d04
US
Los Angeles ·
us-west-1
Score
2,284
FP16 TFLOPS
2,896
vCPU · RAM
128 · 1024G
NVMe disk
500G
5,000 MB/s
Network ↓ ↑
10 Gbps · 5 Gbps
Open ports
100
Reliability
98.59%
$4.928
/hr
≈ $3,603/mo (incl. 64 GB disk)
Market price
$7.040
/hr
Deploy now
Billed per second
1× A100 PCIe
40 GB HBM2 / GPU · AMD EPYC 7763 · PCIe 4.0 · CUDA 13.0 ·
m-02f359f6
SG
Singapore ·
ap-southeast-1
Score
247
FP16 TFLOPS
312
vCPU · RAM
24 · 64G
NVMe disk
4,000G
3,200 MB/s
Network ↓ ↑
2.5 Gbps · 500 Mbps
Open ports
64
Reliability
99.14%
$0.553
/hr
≈ $409/mo (incl. 64 GB disk)
Market price
$0.790
/hr
Deploy now
Billed per second
8× A100 PCIe
40 GB HBM2 / GPU · AMD EPYC 7543 · PCIe 4.0 · CUDA 12.4 ·
m-4390db1c
GB
London ·
eu-west-2
Score
1,966
FP16 TFLOPS
2,496
vCPU · RAM
128 · 1024G
NVMe disk
1,000G
3,200 MB/s
Network ↓ ↑
10 Gbps · 5 Gbps
Open ports
32
Reliability
98.48%
$4.424
/hr
≈ $3,235/mo (incl. 64 GB disk)
Market price
$6.320
/hr
Deploy now
Billed per second
8× H100 PCIe
80 GB HBM2e / GPU · AMD EPYC 9654 · PCIe 5.0 · CUDA 13.0 · NVLink ·
m-15fba106
SE
Stockholm ·
eu-north-1
Score
4,774
FP16 TFLOPS
6,048
vCPU · RAM
64 · 1024G
NVMe disk
2,000G
3,200 MB/s
Network ↓ ↑
2.5 Gbps · 5 Gbps
Open ports
16
Reliability
98.67%
$11.368
/hr
≈ $8,304/mo (incl. 64 GB disk)
Market price
$16.240
/hr
Deploy now
Billed per second
2× H100 PCIe
80 GB HBM2e / GPU · Intel Xeon Platinum 8480+ · PCIe 5.0 · CUDA 12.8 ·
m-7a1b1af5
US
Dallas ·
us-central-1
Score
1,189
FP16 TFLOPS
1,512
vCPU · RAM
48 · 256G
NVMe disk
4,000G
5,000 MB/s
Network ↓ ↑
2.5 Gbps · 1 Gbps
Open ports
64
Reliability
98.32%
$2.842
/hr
≈ $2,080/mo (incl. 64 GB disk)
Market price
$4.060
/hr
Deploy now
Billed per second
1× H100 PCIe
80 GB HBM2e / GPU · Intel Xeon Platinum 8480+ · PCIe 5.0 · CUDA 12.4 ·
m-176bae12
US
Virginia ·
us-east-1
Score
594
FP16 TFLOPS
756
vCPU · RAM
12 · 32G
NVMe disk
500G
6,800 MB/s
Network ↓ ↑
5 Gbps · 1 Gbps
Open ports
32
Reliability
98.22%
$1.421
/hr
≈ $1,042/mo (incl. 64 GB disk)
Market price
$2.030
/hr
Deploy now
Billed per second
1× A100 SXM4
80 GB HBM2e / GPU · AMD EPYC 7543 · PCIe 4.0 · CUDA 13.0 ·
m-75674aa4
HK
Hong Kong ·
ap-east-1
Score
249
FP16 TFLOPS
312
vCPU · RAM
12 · 32G
NVMe disk
500G
5,000 MB/s
Network ↓ ↑
10 Gbps · 1 Gbps
Open ports
32
Reliability
99.94%
$0.693
/hr
≈ $511/mo (incl. 64 GB disk)
Market price
$0.990
/hr
Deploy now
Billed per second
1× A100 SXM4
80 GB HBM2e / GPU · AMD EPYC 7543 · PCIe 4.0 · CUDA 12.4 ·
m-6e9d11f3
GB
London ·
eu-west-2
Score
245
FP16 TFLOPS
312
vCPU · RAM
24 · 48G
NVMe disk
4,000G
2,800 MB/s
Network ↓ ↑
2.5 Gbps · 5 Gbps
Open ports
32
Reliability
98.29%
$0.693
/hr
≈ $511/mo (incl. 64 GB disk)
Market price
$0.990
/hr
Deploy now
Billed per second
4× A100 SXM4
80 GB HBM2e / GPU · Intel Xeon Gold 6348 · PCIe 4.0 · CUDA 13.0 ·
m-31e0b6de
US
Dallas ·
us-central-1
Score
989
FP16 TFLOPS
1,248
vCPU · RAM
96 · 128G
NVMe disk
2,000G
3,200 MB/s
Network ↓ ↑
2.5 Gbps · 500 Mbps
Open ports
100
Reliability
99.05%
$2.772
/hr
≈ $2,029/mo (incl. 64 GB disk)
Market price
$3.960
/hr
Deploy now
Billed per second
8× A100 SXM4
80 GB HBM2e / GPU · AMD EPYC 7763 · PCIe 4.0 · CUDA 12.8 ·
m-3fac5b0c
HK
Hong Kong ·
ap-east-1
Score
1,991
FP16 TFLOPS
2,496
vCPU · RAM
64 · 256G
NVMe disk
8,000G
6,800 MB/s
Network ↓ ↑
5 Gbps · 1 Gbps
Open ports
64
Reliability
99.69%
$5.544
/hr
≈ $4,052/mo (incl. 64 GB disk)
Market price
$7.920
/hr
Deploy now
Billed per second
Show 30 more
All machines are demo data; the same GPU model is priced the same on every host.