Pricing, usage limits, and hosted model offering are very unclear

[sagirogluata]

I’ve been evaluating Continue and I really like the overall idea: open-source IDE integrations, support for local models, and an opinionated Hub/CLI story for “continuous AI” workflows. But as a potential paying user, I’m really confused about what I’d actually be paying for, especially if I want to use your hosted models instead of only my own API keys or local models.

“$10/dev/mo covers config sync, secrets proxy, governance, etc. All model costs still come from your own providers unless you buy the Models Add-On / CLI credits.”

Right now it feels like I’m paying for “Team” mostly to share things I could already share via Git, YAML, or a GitHub repo+docs.

What is the exact pricing structure for the Models Add-On?

What are the concrete usage limits?

How many chat / agent / autocomplete requests per month does the Add-On include?

Are there daily / weekly throttles or only a monthly cap?

“Add-on” and “top-up” feel like they imply that baseline limits are insufficient.

[bdougie]

Hi @sagirogluata the pricing for Models addon has move to a credit model which gets you access to hosted models. You pay for credit used which is equivalent to the model's token cost (at cost). We have been heads down working on a Cloud Agent platform which will also use the same models and credit usage model.

For most users $20 per month sufficient. For those using large models or have a team using the same model-addon, top-up is necessary if the team goes over.

We do not throttle, but provide a message in the response when you run out.

We do not count agent reqeusts