Basic

$0
+ compute+ resource
/ month

Perfect for individuals and small teams to get started.

GET STARTED

No credit card required

  • No subscription fee
  • Up to 48 CPUs + 2 GPUs concurrently

Standard

$30
+ compute+ resource
/ month

Designed to meet the needs of collaborative teams and growing businesses.

SIGN IN TO UPGRADE

  • Multi-user support for collaboration
  • Custom runtime environments
  • Dedicated account manager
  • Up to 192 CPUs + 8 GPUs concurrently
  • Unlimited rate Model APIs

Enterprise

Custom

Built for organizations needing a high level of SLAs, performance, and compliance.

CONTACT US

Standard features plus

  • Custom integration and support
  • Self-hosted deployments
  • Dedicated API support for control plane
  • Audit log and RBAC (coming soon)
  • Prioritize your requests on our roadmap
Compute costs
TypeNameCPUMEMGPUGPU Memory
CPU
cpu.small14 GB--$0.0495
cpu.medium28 GB--$0.099
cpu.large416 GB--$0.198
NVIDIA-A10
gpu.a10832 GB1 × NVIDIA-A1024 GB$1.212
gpu.a10.6xlarge2496 GB1 × NVIDIA-A1024 GB$1.212
NVIDIA-A100
gpu.a100-40gb12192 GB1 × NVIDIA-A10040 GB$3.048
gpu.a100-80gb12192 GB1 × NVIDIA-A100-80GB80 GB$3.21
gpu.2xa100-40gb24384 GB2 × NVIDIA-A10080 GB$6.096
gpu.2xa100-80gb24384 GB2 × NVIDIA-A100-80GB160 GB$6.42
gpu.4xa100-40gb48768 GB4 × NVIDIA-A100160 GB$12.18
gpu.4xa100-80gb48768 GB4 × NVIDIA-A100-80GB320 GB$12.84
gpu.8xa100-40gb961536 GB8 × NVIDIA-A100320 GB$24.36
gpu.8xa100-80gb961536 GB8 × NVIDIA-A100-80GB640 GB$25.68
NVIDIA-A6000
gpu.a6000864 GB1 × NVIDIA-A600048 GB$1.65
NVIDIA-H100
gpu.h100-pcie12192 GB1 × NVIDIA-H100-PCIe80 GB$3.9
gpu.h100-sxm12192 GB1 × NVIDIA-H100-80GB-HBM380 GB$4.2
gpu.2xh100-sxm24384 GB2 × NVIDIA-H100-80GB-HBM3160 GB$8.4
gpu.4xh100-sxm48768 GB4 × NVIDIA-H100-80GB-HBM3320 GB$16.8
gpu.8xh100-sxm961536 GB8 × NVIDIA-H100-80GB-HBM3640 GB$33.6
Model API costs
ModelPrice
WizardLM-2 7B$0.07 / million tokens
WizardLM-2 8x22B$1 / million tokens
Toppy M 7B$0.07 / million tokens
OpenChat 3.5$0.07 / million tokens
Nous: Hermes 13B$0.18 / million tokens
MythoMax L2 13b$0.18 / million tokens
Mixtral 8x7b$0.5 / million tokens
Mixtral 8x22b$0.8 / million tokens
Mistral 7B$0.07 / million tokens
Llama3 8b$0.07 / million tokens
Llama2 13b$0.3 / million tokens
Llama3 70b$0.8 / million tokens
Gemma 7b$0.07 / million tokens
Dolphin Mixtral 8x7b$0.5 / million tokens
Lepton Search (beta)$0.015 / step
Stable Diffusion XL$0.00015 / step
Stable Video Diffusion$0.0092 / step
Super Resolution$0.00015 / image
WhisperX$0.00007 / second
Basic
Standard
Enterprise
Features
Subscription fee
$0 / month
$30 / month
Contact Us
Free credits
$10
$10
Contact Us
Storage
$0.153 / GB / Month
$0.153 / GB / Month
$0.153 / GB / Month
Quota limit
48 CPU, 2 GPU, 1 Queue, 1 KV
192 CPU, 8 GPU, 10 Queue, 10 KV
Unlimited
Model APIs rate limit
10 requests per minute
Unlimited
Unlimited
Multi-user support
Elevated quota for scaling
Custom runtime environments
Dedicated account manager
Custom integration and support
Self-hosted deployments
Dedicated API support for control plane
Audit log and RBAC (coming soon)
Tuna
Training
Free for now
Free for now
Free for now
Training models
No Limit
No Limit
No Limit
Dedicated inference
$3.6 / hour
$3.6 / hour
$3.6 / hour