M

Modular Cloud

Infrastructure service

by Modular

Modular's hosted inference service — shared and dedicated frontier-model endpoints, and the same stack deployed into a customer's own VPC.

Find alternatives to Modular Cloud

Three editions on the pricing page: Modular Cloud billed 'PER TOKEN OR GPU HOUR', Bring Your Own Cloud billed 'PAY PER MINUTE', and Enterprise as a custom engagement. Sample published endpoint rates include DeepSeek V4 at $1.74 input / $3.48 output per 1M tokens and FLUX.2-dev at $10 per 1K tokens.

What it does

• Model inference
• Model hosting

Sources

Pricing

Pricing page · 18 Sep 2026

Rates and billing units quoted from the page.

Description

Official documentation · 18 Sep 2026

Editions, billing units and sample endpoint rates read from Modular's own pricing page, 2026-09-18; docs.modular.com's own heading for it is 'Deploy with Modular Cloud — A simple and powerful inference solution.'

Sold within

Pricing page · 18 Sep 2026

Bought directly as metered inference, not as an add-on to MAX.

URL

Pricing page · 18 Sep 2026

Fetched 2026-09-18, 200. Modular publishes no standalone marketing page for the cloud service — modular.com/platform returns 404 — so the pricing page, which names the editions, is the record's own page. Worth repointing if Modular ships one.

Also from Modular

• MAX Infrastructure service

  Open-source AI serving and modelling framework: a Python API, model pipelines and GPU kernels for NVIDIA, AMD and Apple hardware.

• Mojo Developer tool

  Apache-2.0 systems language for writing fast code across CPUs, GPUs and other accelerators without vendor lock-in, published by Modular.

Open Modular in the directory

Something wrong here? Send a correction — quote this product id: modular-cloud.