Keera Code
A coding agent that stays inside your perimeter
Plan, change across file boundaries, run the test suite, open the merge request. In the terminal or in the IDE - and every action belongs to a person with a name.
What Keera Code is
A set of open-weight models that write code, run on Swiss GPUs or in your own data centre. On top of them sits an ordinary API: whatever talks to OpenAI today talks to Keera tomorrow.
The API
OpenAI-compatible. One base URL, one key, and any client that already speaks to OpenAI speaks to Keera - no new SDK, no plugin.
The models
Open weights, tuned for code. You can read what is inside them, and download them and take them with you whenever you like.
The inference server
vLLM and llm-d spread the requests across many GPUs. That keeps answers fast even when several teams are working at once.
The place
Swiss data centres - or your own racks, with no internet at all if that is what you need.
The models
Open-weight models, tuned for code, run on Swiss GPUs. Pick one per workspace - or let the gateway route: small changes to the fast model, hard reasoning to the strong one.
| Model | Context | Best for |
|---|---|---|
| Keera Deep | 256k | Agentic work across many files |
| Keera Swiss | 128k | Code and documentation in German, French and Italian |
| Keera Fast | 64k | Completion, tests, commit hygiene |
| Your own fine-tune | - | Internal frameworks and legacy code |
Keera Deep and Keera Fast are derived from Qwen-Coder, Keera Swiss from Apertus. All three are Apache-2.0: you can download the weights and take them with you.
Hosted providers - today the OpenAI models - sit behind the same policies, but run on the provider's infrastructure and not on Swiss GPUs. They are blocked by default; a tenant opens them per team and data class, or not at all. Checked September 2026.
Where your engineers already work
CLI, VS Code, JetBrains and headless in the pipeline - all through one OpenAI-compatible base URL. Your MCP servers, internal APIs and tools are declared per project in AGENTS.md.
Every token Keera Code spends runs through Keera Gateway: the same policies and the same audit log as everywhere else.
Try Keera
We onboard one team, connect it to your repos and hand you the numbers. After 30 days you decide.