The Ichnus model family
One assistant, three tiers. Every model in the Ichnus family is free for everyone, tuned for a different job, and held to the same published evaluation standard before release.
Fast
Krus
Built for everyday questions, fast replies, and running at scale without the wait. Krus handles the small stuff so the bigger models stay available for the hard stuff.
- Quick lookups and factual questions
- Drafting and editing
- Simple chat and customer-facing flows
Balanced
Mantus
Strong reasoning at a comfortable speed. Mantus is the default across Ichnus — the model most people meet first, and the one most people settle into.
- Multi-step reasoning and planning
- Writing, analysis, and research
- Long-document work
Frontier
Corus
Built for hard problems — deep research, complex reasoning, and difficult judgment calls. Corus is available free to everyone, with fair-use limits so it stays available.
- Deep research and synthesis
- Complex, multi-step reasoning
- Code, math, and high-stakes decisions
How to choose
If you just want help with something quick, start with Krus — it is fast and costs almost nothing to serve. For everyday work — writing, analysis, planning — Mantus is the default. When the problem is genuinely hard, switch to Corus and give it time to think.
Ichnus chooses the right tier for your question automatically. When you need to decide for yourself, you can switch at any time in the app.
One standard for safety
Every model — including the fast one — passes the same evaluation suite before release. Being cheap to run is not a license to be less careful. You can read about the standard in our approach to model evaluations.
Benchmarks
Each model page includes a benchmark table with its scores on public benchmarks and on our internal evaluations. The figures are illustrative, as everything on this site is — Curos is a concept project.