Cerebras()
Open-weight models at very high throughput, with your own API key. Costs go direct to the vendor at their rates — no markup, nothing on your Axon balance.
export default defineProfile({
providers: [Cerebras()],
})
Serves a fixed list — see below.
Credentials
Set the environment variable, or pass key on the declaration:
CEREBRAS_API_KEY=...
providers: [Cerebras({ key: process.env.MY_OWN_VAR })]
Unlike OpenRouter() and
Codex(), direct providers are not vaulted yet: the
credential comes from the agent's environment. A deployed agent needs it set as a runtime
secret, or should use a vaulted route instead.
Declaring this provider without the variable set is not an error. The failure comes when something tries to use it, and names the variable you need.
Options
Cerebras(options?: ProviderOptions)
| Option | Default | Description |
|---|---|---|
key | CEREBRAS_API_KEY | Your own credential, passed directly |
url | the provider's API | Endpoint override |
slots | unbounded | Ceiling on concurrent calls a fanned-out role may spend here |
An inference host, not a vendor
Like Groq(), Cerebras serves other vendors' open weights
under its own names, so this provider serves an explicit list rather than deriving
them. Currently mapped: GPT-OSS 120B and Gemma 4 31B.
A model absent from the list is not offered rather than guessed at — see Groq for why that is the right failure.
Model ids
Cerebras's own names:
export default defineAgent({ model: "cerebras:gpt-oss-120b" })