CAMELSTREAM · AGENT GUIDE
Unlimited frontier intelligence for always-on OpenClaw.
OpenClaw keeps agents running around the clock, which makes per-token pricing a gamble. Route it through camelStream and the inference bill is $5 a month per stream, no matter what your automations get up to overnight.
Setup
Register the provider.
Add the API key
Expose CAMEL_API_KEY to the OpenClaw gateway process using your existing secret-management approach.
Register camelStream
Merge the custom provider shown here into the OpenClaw configuration and set camel-stream/auto as the primary model.
Verify a bounded workflow
Restart the gateway and test one low-risk agent workflow. Confirm generation, tool use, and any channel delivery before enabling an always-on automation.
{
agents: {
defaults: {
model: { primary: "camel-stream/auto" }
}
},
models: {
mode: "merge",
providers: {
"camel-stream": {
baseUrl: "https://stream.camelai.com/v1",
apiKey: "${CAMEL_API_KEY}",
api: "openai-completions",
models: [
{
id: "auto",
name: "Auto",
contextWindow: 262144
}
]
}
}
}
}stream.camelai.com/v1
auto
openai-completions
env CAMEL_API_KEY
Why OpenClaw on Stream
An always-on agent needs an always-on price.
OpenClaw agents don't keep office hours. They watch channels, run scheduled jobs, and wake up whenever something needs handling, which makes their token usage spiky and hard to forecast. Stream flattens that curve: however busy the night was, the inference line on your bill reads the same in the morning.
Budget the unpredictable
Persistent agents generate traffic at unpredictable times. A fixed monthly price per stream makes the inference layer boring, in the best possible way.
The gateway stays in charge
OpenClaw keeps orchestration, channels, and tool policy. camelStream just answers the model calls, over the OpenAI-compatible API the gateway already speaks.
Primary or failover
Run Stream as the primary provider, or register it as a fallback in an OpenClaw failover design once you've validated the behavior you want.
The deal
What every stream includes.
Frontier models only
You never pick a model; keeping the fleet smart is our job. Every model we serve scores 70% or higher on Terminal-Bench 2.1, or 50 or higher on the Artificial Analysis Intelligence Index, and always at its newest public release.
260K context, guaranteed
Every request gets at least a 260K-token window, and more when the model serving you supports it. Beyond the guaranteed window we compact the middle of the conversation first; your original task and latest turns stay intact.
One generation per stream
Unlimited works because each stream runs one generation at a time. Extra requests wait in a visible queue instead of raising your bill; add streams for parallel capacity.
$5 a month, flat
No token metering, no overage charges, cancel anytime. Sourcing efficiently across the fleet is what keeps the price flat.
The fleet changes as new models launch. The floor doesn't. The full spec sheet lives on the Stream page, and speed numbers are published targets, not a service-level agreement — the fine print is in the terms.
Troubleshooting
Check these first.
Most connection issues come down to the endpoint, API mode, model ID, or key. Start here before anything else.
Set api to openai-completions and use the base URL ending in /v1.
The primary-model reference must include both the provider and model: camel-stream/auto.
Restart the OpenClaw gateway after changing provider configuration or environment variables.
Test channel automations with a low-risk task before granting broader tool or filesystem access.
Frequently asked questions
OpenClaw, answered.
More guides
Point another agent at Stream.
Give OpenClaw the whole stream.
Unlimited frontier intelligence over the API format OpenClaw already speaks, for $5 a month per stream.
Get your API key