Pricing

Choose the right product for your stage

Pipeline catalyst

Our most cost-effective offering. Pay only per seconds of compute of any AI model.

Pipeline core

Our most secure, scalable and fast offering to deploy ML models. Enterprise-grade platform to deploy machine learning anywhere.

Catalyst

Core

Pricing

Compute cost

Pay per second, see table below

Custom

Account cost

Free ($0 per month)

Custom

Egress cost

Free ($0 per GB/month)

Custom

Storage cost

Free ($0 per GB/month)

Custom

Features

Access to public pipelines

Yes

Yes

Deploy private or public pipelines

Yes

Yes

Custom environments

Yes

Yes

Queue jobs

Yes

Yes

Versioning

Yes

Yes

Auto-GPU fragmentalization

Yes

Yes

P75 System Latency

35ms

25ms

P95 System Latency

75ms

40ms

Infrastructure monitoring

No

Yes

Private VPC

No

Yes

Deploy on GCP, AWS, Azure, hybrid or on-prem

No

Yes

Support

Support

Email & Discord

Slack channel

Per second

Per hour

CPU

CPU

24 GB RAM

$0.000028

$0.10

GPU

1x Nvidia T4

16GB total GPU VRAM

$0.0000694

$0.25

1x Nvidia A100 40GB

40GB total GPU VRAM

$0.000556

$2

2x Nvidia A100 40GB

80GB total GPU VRAM

$0.00139

$5

4x Nvidia A100 40GB

160GB total GPU VRAM

$0.00278

$10

1x Nvidia A100 80GB

80GB total GPU VRAM

$0.000694

$2.50

2x Nvidia A100 80GB

160GB total GPU VRAM

$0.00167

$6

4x Nvidia A100 80GB

320GB total GPU VRAM

$0.00333

$12

1x Nvidia V100 16GB

16GB total GPU VRAM

$0.000347

$1.25

1x Nvidia V100 32GB

32GB total GPU VRAM

$0.000417

$1.50

1x Nvidia L4

24GB total GPU VRAM

$0.000139

$0.50

1x Nvidia A5000

24GB total GPU VRAM

$0.000139

$0.50

Experience the power of instant engineering with Mystic

Book a demo