Cerebrium

Automated

Serverless GPU inference platform for AI/ML with pay-per-second billing, sub-second cold starts, and autoscaling from zero.

Visit website Pricing page

Active plans

Total changes

Last change

Jun 2, 2026

Price trend

no history

What we know about Cerebrium’s pricing

Moderately active

Changes in 12 months

Cadence

Multiple changes per day

Direction

New plan recently added

Category position

Pricier than 87% of AI & ML tools (2268 compared)

Current pricing

Hobby

usage based

Free

usage based$0(Free + compute)
monthly$0/mo(Free + compute / month)

3 user seats
Up to 3 deployed apps
500 CPU containers, 5 concurrent GPUs
1-day log retention
Pay-per-second compute billing
Community support

Standard

monthly

$100/mo

$100/mo + compute

User seats(Unlimited)
Deployed apps(Unlimited)
Containers(1000)
Concurrent GPUs(30)
Custom domains
Log retention(30 days)
Support(Private Slack Channel)

Enterprise

custom

Custom

Volume discounts
Unlimited concurrent GPUs
White-glove onboarding
ML engineering services
SOC2, HIPAA, GDPR, ISO 27001

Price history

Last 3 changes

Date	Plan	Change	Old	New	Δ%
Jun 2, 2026	Enterprise	new plan	—	—	—
Jun 2, 2026	Standard	new plan	—	$100	—
Jun 2, 2026	Hobby	new plan	—	$0	—

Similar to Cerebrium.

No confirmed competitors yet. These are AI & ML products at a similar price, matched automatically — not verified alternatives.

E2B

From $150/mo + usage

Cloud sandbox infrastructure for running AI-generated code securely, with millisecond spin-up and pay-per-second billing.

Inferless

From $0.000555/sec

Serverless GPU platform for deploying ML models with pay-per-second pricing.

Novita AI

From Usage-based per model (e.g. DeepSeek V4 Pro: $1.60/1M in)

AI inference API with 100+ models including LLM, image, audio, and video generation with serverless and dedicated endpoints.

Cerebras Inference

From $10

Ultra-fast AI inference API powered by Cerebras Wafer-Scale processors, delivering the world's fastest LLM inference speeds.

FriendliAI

From $0.10/1M tokens (Llama-3.1-8B-Instruct)

AI inference platform offering model APIs, dedicated endpoints, and containers for fast and affordable LLM inference.

GPTPacker

From $5/24 hours

AI-powered 3D container loading and packing optimizer that generates optimal loading plans from plain English descriptions in seconds.

PriceTrack