Quickstart

1. Get an invite

Join the waitlist at inferam.ai. We approve invites by hand. When yours is approved, you get an email with a sign-in link.

2. Install the CLI

On macOS or Linux (Apple silicon, ARM or x86-64; on Windows, use WSL):

curl -fsSL https://inferam.ai/install | sh

The installer puts inferam in ~/.local/bin without root, after checking the download against its published SHA-256 checksum. If that folder isn’t on your PATH, the installer tells you what to add.

inferam auth login

Your browser opens a page showing a code. Check that it matches the code in your terminal, then approve it. If you’re on a machine without a browser, such as over SSH, open the printed URL on any device instead. The CLI stores a device key in your OS keychain (or in a file only you can read). It’s named “CLI on <hostname>” and is listed on inferam.ai/keys.

4. Run your agent

inferam claude      # or: inferam dsh, inferam codex, inferam opencode, inferam pi

The agent runs on Inferam for that session only. Your agent’s own config files aren’t touched, and any flags you add pass straight through. If the agent isn’t installed, the CLI offers to install it with the agent’s official installer.

When you quit, the CLI prints the command that resumes the session on Inferam, for example:

Resume this session on Inferam: inferam claude --resume 096a011e-54e2-4c60-a3e8-c2d8b1199c67

5. Call the API directly

Create a key on inferam.ai/keys. The full key is shown once, so copy it then. Then:

export INFERAM_API_KEY=ifr_...

curl https://api.inferam.ai/v1/chat/completions \
  -H "Authorization: Bearer $INFERAM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "DeepSeek-V4.1-Flash-1017",
    "messages": [{"role": "user", "content": "Write a haiku about type systems."}]
  }'

Official SDKs work by changing only the base URL and key:

from openai import OpenAI

client = OpenAI(base_url="https://api.inferam.ai/v1", api_key="ifr_...")
r = client.chat.completions.create(
    model="DeepSeek-V4.1-Flash-1017",
    messages=[{"role": "user", "content": "Hello"}],
)
print(r.choices[0].message.content)

import anthropic

client = anthropic.Anthropic(base_url="https://api.inferam.ai", api_key="ifr_...")
m = client.messages.create(
    model="DeepSeek-V4.1-Flash-1017",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)
print(m.content[-1].text)

The Anthropic SDK takes the base URL without /v1; the OpenAI SDK takes it with /v1.

See your token usage on inferam.ai/usage. New requests show up within a few minutes.
Read the API overview for limits, the capacity queue and error handling.

Quickstart

1. Get an invite

2. Install the CLI

3. Sign in

4. Run your agent

5. Call the API directly

Next