ServerlessBeta

Deploy a model. Get a URL.

An OpenAI-compatible endpoint for any open model, backed by an autoscaling worker pool. Traffic comes, workers start. Traffic stops, they scale to zero, and so does the GPU bill.

Deploy an endpoint
Pay per minute of worker runtimeOnly file storage when idlevLLM · SGLang · Ollama

Speaks OpenAI.

Point the SDK you already use at your endpoint. Nothing else changes.

Sleeps when traffic does.

Set min workers to 0 and an idle endpoint costs storage, not GPUs.

Any open model.

Name a Hugging Face repo, or an Ollama tag. Gated models take your HF token.

What is a serverless GPU endpoint?

A serverless endpoint is a model with a URL and no machine to think about. You name the model and the GPU; we run a pool of workers that grows when requests queue and shrinks when they stop. At zero traffic the pool can sleep entirely, and while it sleeps you pay for stored weights, not for waiting hardware.

The other two products rent you time on a box. This one rents you an answer to a request. That is why it bills differently: per minute of worker runtime, metered per worker, instead of per minute of a machine you keep.

Frameworks

Three engines, your model.

Bring the model as a Hugging Face repo or an Ollama tag. Custom containers are not part of the beta.

vLLM

Fast LLM inference and serving. The default for chat and completion workloads.

SGLang

Fast serving for large language and vision-language models.

Ollama

Lightweight runner for open LLMs. Name a tag like gemma3:4b and go.

How it works

Four steps, then it answers for itself.

01

Pick a framework and a model

vLLM, SGLang or Ollama; a Hugging Face repo or an Ollama tag.

02

Shape the pool

GPU type, 1 to 8 GPUs per worker, min and max workers. Min 0 means it can sleep.

03

Get the URL

OpenAI-compatible, authenticated with your JarvisLabs API token.

04

Let it breathe

Workers start when requests queue and stop after your idle timeout.

Configuration

Every knob, in the open.

Serverless platforms tend to hide their scaling behavior. Ours is a form you fill in.

KnobRangeWhat it does
GPU typeL4 · H100 · H200 · RTX Pro 6000One type per deployment
GPUs per worker1 · 2 · 4 · 8Size one worker to fit the model
Min workers0 to 1000 = the pool may sleep entirely
Max workers1 to 100The ceiling autoscaling can reach
Concurrent requests1 to 100 (Ollama: 4)Per worker, before the queue grows
Idle timeout0s to 24hHow long a quiet worker stays warm
Wait time0s to 10mHow long a request waits for a worker before failing
File storage50 GB to 1 TBHolds the model weights, billed around the clock

After deploy, name, idle timeout and wait time can change on a running deployment. Changing workers, GPU or model means recreating it.

Billing

The bill breathes with the pool.

While a worker runs

Its GPUs bill per minute of runtime, plus the worker's 50 GB root disk at $0.00014/GB/hour. Each worker is metered on its own.

While the pool sleeps

Nothing runs, so no GPU charges. File storage keeps your weights warm at $0.00014/GB/hour, around the clock.

What you will not find

No request fees, no per-token markup, no minimums. The dashboard shows the live running cost per worker as it accrues.

Observability

Watch it work.

Live logs

Streamed from every worker, filterable per worker, tail included.

Worker state

Healthy counts, queue depth, model-download progress, and each worker's scale-down countdown.

Running cost

Per-worker cost as it accrues, and the deployment total, on the same screen.

Choosing

Serving here. Everything else, next door.

The three products share GPUs and a per-minute meter. They differ in what you are renting.

ServerlessThis pageTemplateOn-Demand VM
Built forServing requestsTraining and notebooksOwning the machine
What you getAn OpenAI-compatible URLA running environment in 1.8sRoot on a box in about 90s
Scales0 to 100 workers, automaticallyOne instance, your callOne machine, your call
BillsPer minute of worker runtimePer minute while runningPer minute while running
Pick this whenyou're serving, not trainingyou want to start training right awaya container is not a machine
Questions

Answers.

Can't find what you're looking for? Reach out to our support team.

The product is young, not gated. You sign up and deploy from the dashboard today. It also means no SLA yet, three frameworks rather than custom containers, and rates that may change as the product settles.

Yes. It serves the chat-completions API, so the OpenAI SDK works by changing the base URL and using your JarvisLabs API token as the key.

A worker cold-starts. The request waits up to your configured wait time (as long as 10 minutes) for a worker to pick it up, then fails if none does. Large models take longer because the weights load first; the dashboard shows download progress.

With min workers at 0, only file storage, at $0.00014/GB/hour. Workers bill only while running.

L4, H100, H200, RTX Pro 6000, at 1, 2, 4, 8 GPUs per worker.

Yes. Add your Hugging Face token as an environment variable on the deployment and name the repo as usual.

Name, idle timeout and wait time can change on a running deployment. Changing workers, GPU or model means recreating it, which the beta keeps deliberately simple.

India today, in two regions. More regions follow the beta.

Not on Serverless yet. vLLM, SGLang and Ollama are the beta engines. For arbitrary containers, an On-Demand VM is the right box.

Get started

Start building on The AI Cloud.