Models
Caelim.
Caelim is the model family developed by OevuAI.
Our in-house model family. We describe each release only by what it demonstrably does — no benchmark tables we can't reproduce, no parameters we can't verify.
| Model | What it does today | Where to use it |
|---|---|---|
| Caelim-2.1 Turbo | Lightweight fast tier — everyday questions, high concurrency, low latency; the free tier and main call volume. | Web chat · API caelim-2.1-turbo |
| Caelim-2.1 Pro | Deep-reasoning tier — complex tasks, long-form writing, deep analysis. | Web chat · API caelim-2.1-pro |
| Caelim-2.1 Beta | Experimental tier — early access to a 3.8M-token context, at lower stability. | Web chat caelim-2.1-beta |
| Caelim-2.5 | The current mobile-first release. | Android app |
Capability claims on this page are limited to features shipped in our own products. Formal evaluation results will be published on /research/evaluations as runs complete.
Direction
Why "learns as it works"
Developmental approach
We are interested in how capabilities develop over time — models that improve through use, feedback, and successive releases — rather than one-shot claims about general intelligence.
Verifiable by default
Every surface on this site is live: chat now, download now, call the API now. The research section publishes data and methods you can check, including checksums.