Your LLMOps platform, hosted in France
The AI app development studio powered by Agenta, installed and maintained by DINAO. Prompt playground, evaluation, and observability — on our French servers.
What is Agenta?
Agenta is an open-source LLMOps platform that helps engineering and product teams build reliable LLM applications faster, bringing together prompt management, evaluation, and observability in one place. It covers the entire lifecycle of an AI application, from prototype to production.
Its interactive playground allows you to compare prompts side-by-side on your test cases, with support for over 50 models and version management via branches and environments. For quality assurance, you create test sets from production data, playground experiments, or CSV files, then run automatic or custom evaluators. Observability provides tracking of costs, latency, and usage, as well as detailed tracing to debug complex workflows.
Agenta is designed for engineering and product teams looking to stabilize their LLM applications through shared tools and a systematic quality assurance approach. The platform supports self-hosting via Docker.
Host Agenta at DINAO
Resource tiers compatible with Agenta prerequisites (minimum 2 vCPU / 4 Go / 20 Go). Hosted in France, fully managed.
- 2 dedicated vCPU
- 4 Go RAM
- 40 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 4 dedicated vCPU
- 8 Go RAM
- 80 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 8 dedicated vCPU
- 16 Go RAM
- 160 GB NVMe
- Daily backups
- Managed & monitored by DINAO
1 hidden tier(s) (insufficient resources for this app) : Découverte
This application uses AI
The container hosts the application, not the AI engine (which requires dedicated GPUs) : inference is handled externally, with your own provider key. Prioritize a sovereign engine — Mistral AI, NumSpot, Scaleway, or OVHcloud AI Endpoints (France, GDPR) ; an international provider (OpenAI, Anthropic…) only if a specific capability requires it. Inference subscriptions are not included in hosting.
Technical details
You might be wondering…
Are my prompts and test data confidential?
Yes. Your prompts, test sets, and traces remain on your instance. In local model mode, nothing leaves. If you evaluate models via external API, the corresponding requests are processed by that provider.
Which models can I connect?
Agenta is compatible with over 50 models: locally hosted open-weight models (sovereign) or external APIs via your own key. You can compare and switch on demand.
Where is the data hosted?
On DINAO infrastructure in France, in one of the available data centers. Your prompts and traces do not leave the territory in local mode.
Do I need technical skills?
The playground and evaluation are accessible to product teams without coding. Developers also have access to an API and SDK. DINAO manages installation and updates.
Can I change plans or export my data?
Yes. You can upgrade or downgrade at any time, and your prompts, test sets, and traces remain exportable — no proprietary lock-in.
