Modal

Serverless cloud platform for running Python code, GPU jobs and sandboxes without managing infrastructure.

Recommended for serverless GPU workloads, with a well-designed SDK and faster setup than provisioning instances by hand,123 and listed among sandbox providers for running AI agents.45 Complaints are isolated: one commenter says its gVisor sandboxes make running Docker inside harder,6 and another doubts that GPU-as-a-service middlemen like Modal are sustainable.7
The HN verdict, written from 7 cited comments
6recommend
1warn against
Nov 17, 2025first mention
Aug 31, 2026latest mention
Nov 25 – Oct 26
Used forfor GPU workloadsfor ML trainingfor LLM inferencefor AI agent sandboxesfor light usersprovisioning ml training instancesinference hosting

Why people recommend it

GPU serverless · 3

Praised for getting function-as-a-service for GPU workloads right and for cutting slow instance setup.

“I love modal. I think they got FaaS for GPU exactly right”

“Modal significantly improves this. Highly recommend.”

Show 1 more

“hooked it up to vLLM on Modal”

Agent sandboxes · 2

Named alongside E2B and others as a good hosted sandbox option for AI agents.

“We use e2b.dev, but Modal is great too.”

“Daytona, E2B, OpenComputer, Freestyle, Blaxel, Vercel, Modal, Cloudflare, Tensorlake, Superserve, etc. etc.”

SDK and CLI · 2

Commenters like its SDK abstractions and say the CLI works well for managing jobs and assets.

“I love modal. I think they got FaaS for GPU exactly right”

“I recommend the service but like I said I haven't tried any others.”

light user, $20-40/monthon “Keeping 20k GPUs Healthy”

Reasonable pricing · 1

A light user found the price reasonable at $20 to $40 a month on top of free credits.

“I recommend the service but like I said I haven't tried any others.”

light user, $20-40/monthon “Keeping 20k GPUs Healthy”

Caveats and dissent

Docker in sandboxes · 1

Its sandboxes use gVisor rather than Firecracker, which makes running Docker inside them harder.

“Modal, but they use gVisor and not Firecracker, which means it's harder for users to run docker inside the box”

Business sustainability · 1

One commenter doubts GPU-as-a-service middlemen reselling public clouds can be profitable long term.

“business model of GPU-as-a-service middleman companies like modal deeply unsustainable”

Often compared with

Who recommends it in their own words

All comments 11

“Modal significantly improves this. Highly recommend.”

“hooked it up to vLLM on Modal”

“Modal, but they use gVisor and not Firecracker, which means it's harder for users to run docker inside the box”

“Modal is awesome, btw”

“Daytona, E2B, OpenComputer, Freestyle, Blaxel, Vercel, Modal, Cloudflare, Tensorlake, Superserve, etc. etc.”

“I love modal. I think they got FaaS for GPU exactly right”

“across Replicate, Modal, and custom GPU clusters”

“I recommend the service but like I said I haven't tried any others.”

light user, $20-40/monthon “Keeping 20k GPUs Healthy”

“business model of GPU-as-a-service middleman companies like modal deeply unsustainable”

“We use e2b.dev, but Modal is great too.”

Threads where it came up

Data: modal.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.

Keyboard shortcuts
j / kNext / previous item
o or EnterOpen selected item
/Search
g hHome
g tTop
g rRising
g cCategories
tToggle dark mode
?This help