CAMELSTREAM · AGENT GUIDE
Wire OpenCode into unlimited frontier intelligence.
OpenCode treats model providers as plain JSON config. Add camelStream as one and every session runs on frontier models for a flat $5 a month per stream, however many tokens it takes.
Setup
One block in opencode.json.
Save your Stream key
Set CAMEL_API_KEY in the shell that launches OpenCode so the configuration can read the credential without committing it to your repository.
Add the custom provider
Create or update opencode.json in your project with the camelStream provider and model configuration shown here.
Select the model
Start OpenCode, open /models, choose camelStream, and select Auto. Test it on a small repository task before moving a long-running session over.
{
"$schema": "https://opencode.ai/config.json",
"provider": {
"camel-stream": {
"npm": "@ai-sdk/openai-compatible",
"name": "camelStream",
"options": {
"baseURL": "https://stream.camelai.com/v1",
"apiKey": "{env:CAMEL_API_KEY}"
},
"models": {
"auto": {
"name": "Auto"
}
}
}
}
}stream.camelai.com/v1
auto
Chat Completions
env CAMEL_API_KEY
Why OpenCode on Stream
A provider you can check in.
OpenCode keeps providers in a JSON file, which means your model setup is code: reviewable, reproducible, and shared with the whole team the moment it lands in the repo. camelStream fits that shape well. The provider block below is the entire integration, and the key stays in an environment variable where it belongs.
Terminal-native, meter-free
Keep OpenCode's workflow, slash commands, and tool loop. The only thing that changes is that generation stops being billed by the token.
Reproducible across repos
Drop the same provider block into any project's opencode.json and every machine resolves the same setup. New teammate, same two-minute onboarding.
Room for agent traffic
Repository exploration and iterative tool use generate more tokens than people expect. On a flat rate, that stops being a reason to keep sessions short.
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.
Make sure the provider ID in opencode.json is camel-stream and the API key environment variable is available to the OpenCode process.
Use @ai-sdk/openai-compatible for the chat-completions endpoint.
If the model does not appear, restart OpenCode and run /models again after saving opencode.json.
Confirm the model ID is auto; Stream picks the frontier model behind it.
Frequently asked questions
OpenCode, answered.
More guides
Point another agent at Stream.
Give OpenCode the whole stream.
Unlimited frontier intelligence over the API format OpenCode already speaks, for $5 a month per stream.
Get your API key