Links

Plans and pricing

Discover usage-based pricing for self-serve model deployment and our Enterprise plan.
With usage-based pricing on the Startup plan, you only pay for the time your model is actively deploying, scaling up or down, or making predictions.

CPU-only instances

Type
vCPU
RAM
Cost/minute
1×2
1
2 GiB
$0.00096
1×4
1
4 GiB
$0.00144
2×8
2
8 GiB
$0.00288
4×16
4
16 GiB
$0.00576
8×32
8
32 GiB
$0.01152
16×64
16
64 GiB
$0.02304
GiB = 2^30 bytes

GPU instances

Add a GPU to accelerate your model inference.
Type
GPU
VRAM
vCPU
RAM
Cost/minute
T4x4x16
NVIDIA T4
16 GiB
4
16 GiB
$0.01753
T4x8x32
NVIDIA T4
16 GiB
8
32 GiB
$0.02507
T4x16x64
NVIDIA T4
16 GiB
16
64 GiB
$0.04013
A10Gx4x16
Nvidia A10
24 GiB
4
16 GiB
$0.03353
A10Gx8x32
Nvidia A10
24 GiB
8
32 GiB
$0.04040
A10Gx16x64
Nvidia A10
24 GiB
16
64 GiB
$0.05413
V100x8x61
Nvidia V100
16 GiB
16
61 GiB
$0.10200
A100x12x144
Nvidia A100
80 GiB
12
144 GiB
$0.17083
GiB = 2^30 bytes

Compare plans

Baseten offers two plans, Startup and Enterprise.
»
Startup
Enterprise
Workspace
Users
5
Unlimited
Role-based access control
Model deployment
Models and versions
Unlimited
Unlimited
Model performance metrics
Version management
Draft models
Model resources
Volume discounts available
Autoscaling
Applications
Applications
3
Unlimited
Public sharing
CRON scheduling
Draft environments
GitHub sync
Import external code
Data
Data connections
Postgres data tables
Security and privacy
API keys
SOC 2 Type II & HIPAA
Multi-tenant and data segregation
Self-hosted Baseten
Data privacy agreements
Support
Email support
Slack and Zoom support
Dedicated forward-deployed engineer
Custom proof-of-concept