Timbal

What is Timbal?

Timbal is the production AI platform enterprise teams use to build, deploy, and govern agents, workflows, and knowledge bases. Define behavior in code or in Studio, run on the model and provider of your choice, and ship to chat, email, voice, and your product UI from a single runtime.

How is Timbal different from LangChain, Make, or Zapier?

LangChain is a framework. Make and Zapier are no-code automation tools. Timbal is the full production stack: a typed Python framework, a Studio for building visually, a runtime that orchestrates agents and workflows, governance and evals for enterprise rollout, and integrations with the systems your team already uses.

Does Timbal support enterprise deployments?

Yes. Timbal supports sovereign hosting (cloud or on-premise), customer-owned model API keys, role-based access controls, and full run-level observability. Every agent run, tool call, and model invocation is auditable, replayable, and gated through evals before promotion to production.

Is Timbal model agnostic?

Yes. Timbal is model-agnostic. Use OpenAI, Anthropic, Google, Mistral, Meta, or self-hosted models — switch providers per agent, per step, or per tenant. Provider fallbacks are first-class so a single failure does not take down a workflow.

Where does Timbal host data?

Customer data stays in the region you choose. Timbal Cloud runs in the EU by default; on-prem and private-cloud deployments keep all data inside your perimeter. We never train on customer data and we sign Data Processing Agreements aligned to GDPR.

How does pricing work?

Free for individuals, paid plans from €25 per month per seat for teams, and custom pricing for enterprise rollouts that need sovereign hosting, SLAs, or volume usage. Same agents, workflows, and knowledge bases on every plan — you pay for usage and seats, not for the right to use a feature.