YI CLOUD

Model API, sandboxes and VMs.

Call models with the OpenAI or Anthropic SDK, start a sandbox to run commands, or lease a VM and SSH in. Models are billed per token, sandboxes per second, VMs per hour.

DEMO

See it work

Yi CloudExample
import osfrom openai import OpenAIclient = OpenAI(base_url="https://cloud.theyilabs.com/v1",                api_key=os.environ["YI_CLOUD_API_KEY"])r = client.chat.completions.create(model="gemini-4-argon",    messages=[{"role": "user", "content": "Capital of France?"}])

WHAT IT DOES

What Yi Cloud does

OpenAI and Anthropic compatible

Chat completions, responses, embeddings and Anthropic Messages, with streaming. Point the official SDKs at our base URL.

Limits per key

Each key belongs to a project, says what it may be used for, and can have a spend limit and a requests-per-minute limit.

Failover

If a deployment is busy or fails before it answers, the request goes to another deployment of the same model.

Sandboxes

Each sandbox is its own virtual machine, and nothing can connect to it from outside. Run commands and move files through the API. Billed per second, with a 60-second minimum.

VMs

Leased by the hour. Log in with your SSH key from the addresses you list. Stop, start, renew, reboot or reinstall it.

Holds and charges

Before a model call or a resource starts, its largest possible cost is held. The charge is never more than the hold.

HOW TO START

How to start

  1. 01

    Create an API key

    Pick a project and what the key may be used for. Limits are optional.

  2. 02

    Call a model

    Point the OpenAI or Anthropic SDK at our base URL. Only the base URL and the key change.

  3. 03

    Start a sandbox or a VM

    Create it through the API, and delete it when you’re done.

CODE → EVAL → GUARD · CLOUD

Other products

01 · BUILD

Yi Code

Tell it what to change. It edits the code in an isolated sandbox, runs the tests and opens a draft PR for you to review.

Explore Yi Code

02 · BEFORE LAUNCH

Yi Eval

Runs your agent for real in a sandbox, through normal tasks, tool faults and attacks. A version that fails the release gate does not ship.

Explore Yi Eval

03 · AT RUN TIME

Yi Guard

Before an agent calls a tool, runs a command or goes out to the network, Guard checks it against your rules.

Explore Yi Guard

All products

Next step

Want to try it on your own chips?

Book a PoC