transparent pricing
Each Model on Dedicated GPU
Pick the combination that fits your workload — pricing applies when deployed in our cloud. For self-host pricing, contact us to learn more.
10models
8GPU types
$/hrsupportednot available
Price per model / GPU
Hourly rate for each model on each GPU. An ✕ means the model isn’t supported on that GPU.
Qwen-3.5 9B
4-Bit8-Bit16-Bit
RTX 5060 Ti
16GB vRAM
$0.36/hr
RTX 5090
32GB vRAM
$0.78/hr
RTX 6000 Ada
48GB vRAM
$0.16/hr
Best value
RTX PRO 6000
96GB vRAM
$2.20/hr
Nvidia L40S
48GB vRAM
N/A
Nvidia H100
80GB vRAM
N/A
Nvidia H200
141GB vRAM
N/A
Nvidia B200
180GB vRAM
$9.10/hr
Qwen-3.5 27B
4-Bit8-Bit16-Bit
RTX 5060 Ti
16GB vRAM
$0.36/hr
Best value
RTX 5090
32GB vRAM
N/A
RTX 6000 Ada
48GB vRAM
N/A
RTX PRO 6000
96GB vRAM
$2.20/hr
Nvidia L40S
48GB vRAM
$2.26/hr
Nvidia H100
80GB vRAM
$4.70/hr
Nvidia H200
141GB vRAM
$5.40/hr
Nvidia B200
180GB vRAM
$9.10/hr
Llama-3.3 70B Instruct
4-Bit16-Bit
RTX 5060 Ti
16GB vRAM
N/A
RTX 5090
32GB vRAM
N/A
RTX 6000 Ada
48GB vRAM
N/A
RTX PRO 6000
96GB vRAM
$2.20/hr
Best value
Nvidia L40S
48GB vRAM
$2.26/hr
Nvidia H100
80GB vRAM
$4.70/hr
Nvidia H200
141GB vRAM
$5.40/hr
Nvidia B200
180GB vRAM
$9.10/hr
Llama-3.1 8B Instruct
4-Bit8-Bit16-Bit
RTX 5060 Ti
16GB vRAM
$0.36/hr
RTX 5090
32GB vRAM
$0.78/hr
RTX 6000 Ada
48GB vRAM
$0.16/hr
Best value
RTX PRO 6000
96GB vRAM
$2.20/hr
Nvidia L40S
48GB vRAM
$2.26/hr
Nvidia H100
80GB vRAM
N/A
Nvidia H200
141GB vRAM
N/A
Nvidia B200
180GB vRAM
N/A
Gemma-4 31B it
4-Bit8-Bit16-Bit
RTX 5060 Ti
16GB vRAM
N/A
RTX 5090
32GB vRAM
N/A
RTX 6000 Ada
48GB vRAM
$0.16/hr
Best value
RTX PRO 6000
96GB vRAM
$2.20/hr
Nvidia L40S
48GB vRAM
$2.26/hr
Nvidia H100
80GB vRAM
$4.70/hr
Nvidia H200
141GB vRAM
$5.40/hr
Nvidia B200
180GB vRAM
$9.10/hr
Nemotron-3 Nano 30B A3B
4-Bit8-Bit16-Bit
RTX 5060 Ti
16GB vRAM
N/A
RTX 5090
32GB vRAM
N/A
RTX 6000 Ada
48GB vRAM
N/A
RTX PRO 6000
96GB vRAM
$2.20/hr
Best value
Nvidia L40S
48GB vRAM
N/A
Nvidia H100
80GB vRAM
$4.70/hr
Nvidia H200
141GB vRAM
$5.40/hr
Nvidia B200
180GB vRAM
$9.10/hr
Qwen-3.6 27B
4-Bit8-Bit16-Bit
RTX 5060 Ti
16GB vRAM
$0.36/hr
Best value
RTX 5090
32GB vRAM
N/A
RTX 6000 Ada
48GB vRAM
N/A
RTX PRO 6000
96GB vRAM
$2.20/hr
Nvidia L40S
48GB vRAM
$2.26/hr
Nvidia H100
80GB vRAM
$4.70/hr
Nvidia H200
141GB vRAM
$5.40/hr
Nvidia B200
180GB vRAM
$9.10/hr
Ministral-3 14B Instruct
4-Bit8-Bit16-Bit
RTX 5060 Ti
16GB vRAM
$0.36/hr
Best value
RTX 5090
32GB vRAM
N/A
RTX 6000 Ada
48GB vRAM
N/A
RTX PRO 6000
96GB vRAM
$2.20/hr
Nvidia L40S
48GB vRAM
$2.26/hr
Nvidia H100
80GB vRAM
$4.70/hr
Nvidia H200
141GB vRAM
$5.40/hr
Nvidia B200
180GB vRAM
$9.10/hr
Ministral-3 8B Instruct
8-Bit16-Bit
RTX 5060 Ti
16GB vRAM
$0.36/hr
RTX 5090
32GB vRAM
$0.78/hr
RTX 6000 Ada
48GB vRAM
$0.16/hr
Best value
RTX PRO 6000
96GB vRAM
$2.20/hr
Nvidia L40S
48GB vRAM
N/A
Nvidia H100
80GB vRAM
N/A
Nvidia H200
141GB vRAM
N/A
Nvidia B200
180GB vRAM
$9.10/hr
Devstral Small-2 24B Instruct
8-Bit
RTX 5060 Ti
16GB vRAM
N/A
RTX 5090
32GB vRAM
N/A
RTX 6000 Ada
48GB vRAM
N/A
RTX PRO 6000
96GB vRAM
$2.20/hr
Best value
Nvidia L40S
48GB vRAM
$2.26/hr
Nvidia H100
80GB vRAM
$4.70/hr
Nvidia H200
141GB vRAM
$5.40/hr
Nvidia B200
180GB vRAM
$9.10/hr
| Model | RTX 5060 Ti 16GB vRAM | RTX 5090 32GB vRAM | RTX 6000 Ada 48GB vRAM | RTX PRO 6000 96GB vRAM | Nvidia L40S 48GB vRAM | Nvidia H100 80GB vRAM | Nvidia H200 141GB vRAM | Nvidia B200 180GB vRAM |
|---|---|---|---|---|---|---|---|---|
Qwen-3.5 9B 4-Bit8-Bit16-Bit | $0.36/hr | $0.78/hr | $0.16/hr Best value | $2.20/hr | $9.10/hr | |||
Qwen-3.5 27B 4-Bit8-Bit16-Bit | $0.36/hr Best value | $2.20/hr | $2.26/hr | $4.70/hr | $5.40/hr | $9.10/hr | ||
Llama-3.3 70B Instruct 4-Bit16-Bit | $2.20/hr Best value | $2.26/hr | $4.70/hr | $5.40/hr | $9.10/hr | |||
Llama-3.1 8B Instruct 4-Bit8-Bit16-Bit | $0.36/hr | $0.78/hr | $0.16/hr Best value | $2.20/hr | $2.26/hr | |||
Gemma-4 31B it 4-Bit8-Bit16-Bit | $0.16/hr Best value | $2.20/hr | $2.26/hr | $4.70/hr | $5.40/hr | $9.10/hr | ||
Nemotron-3 Nano 30B A3B 4-Bit8-Bit16-Bit | $2.20/hr Best value | $4.70/hr | $5.40/hr | $9.10/hr | ||||
Qwen-3.6 27B 4-Bit8-Bit16-Bit | $0.36/hr Best value | $2.20/hr | $2.26/hr | $4.70/hr | $5.40/hr | $9.10/hr | ||
Ministral-3 14B Instruct 4-Bit8-Bit16-Bit | $0.36/hr Best value | $2.20/hr | $2.26/hr | $4.70/hr | $5.40/hr | $9.10/hr | ||
Ministral-3 8B Instruct 8-Bit16-Bit | $0.36/hr | $0.78/hr | $0.16/hr Best value | $2.20/hr | $9.10/hr | |||
Devstral Small-2 24B Instruct 8-Bit | $2.20/hr Best value | $2.26/hr | $4.70/hr | $5.40/hr | $9.10/hr |
How ourpricing works?
Transparent, Real-Time Pricing
No hidden fees. No surprises—prices update live as you use it. Find the best GPU for the best price.
A Platform Built for Discovery
Find your ideal instance and easily filter by model, memory, region, or price. Find your perfect match in seconds.
Flexible, Fluid Billing
Pay per second, on-demand, enterprise, or reserved capacity—tailored to your workload.
Instant Deployments, Built to Scale
Make it easy—Launch fast, scale on demand, and automate everything with our CLI or API.
Give your AI workloads exactly what they need without overspending, guesswork, overbuying, or long-term lock-ins.
Contact Sales Team