Your OpenAI, Anthropic, or Gemini key. Your invoice. Your DPA.
On any plan, Kelu can call the model on your provider account with your key. You keep the model choice, the usage bill, and the data-processing agreement you already have with OpenAI / Anthropic / Google. Kelu is the platform layer; the LLM traffic runs on you.
How BYOK works
Per-workspace credentials — every customer-facing answer runs on the account you configured
Bring Your Provider Credential
- Configure a provider credential per workspace
- Applies to every knowledge base in the workspace
- Credentials encrypted at rest with AES-256-GCM
- Rotate or revoke in the dashboard; no re-deploy required
- Fallback to the deployment provider when a credential is missing
Coverage Across the Answer Path
- Streaming widget and portal answers — on your credential
- Multi-knowledge-base answers — on your credential
- Slack and Microsoft Teams bots (AnswerForBot) — on your credential
- Form deflector, including grounding verification — on your credential
- Reranking on the LLM grader fallback — on your credential
What Still Runs on Deployment Provider
- Kelu's internal diagnostic tooling
- Conversation title generation (small fixed-cost call)
- Image-to-text (DescribeImage) for chat attachments
- Embeddings, unless you use an OpenAI key on the OpenAI API
- Hosted cross-encoder rerankers (Cohere/Voyage/Jina — not LLM providers)
The two questions Enterprise buyers actually ask
'Whose account is charged for the tokens?' and 'Whose DPA covers the data?' — the two questions every enterprise procurement conversation lands on. BYOK answers both with 'yours'. Traffic runs on your OpenAI / Anthropic / Gemini account under the terms you already negotiated. Kelu is the platform above the model, not the LLM vendor.
- Volume LLM contracts you already negotiated apply directly
- Model choice is yours — no vendor lock-in via the platform
- Data-processing terms are yours — no additional third-party review
- Rotate the credential without re-deploying anything
- Revert to default switches back to Kelu's models at any time
Answer model
Choose which language model generates your answers.
claude-sonnet-4-6on AnthropicBilled to your accountWhere does inference run?
Provider
Model
API keySaved
What runs on your account
- Answersyour account
- Query planningyour account
- Passage scoringKelu— scored by a Kelu model
- EmbeddingsKelu
Private content stays private, and secrets stay sealed
Credentials are encrypted
Connection secrets are sealed with AES-256-GCM. If one can't be opened, the sync stops instead of running without it.
Private stays private
Internal knowledge bases have no public surface at all. To keep past tickets private, add them to an internal knowledge base.
Every answer from your docs shows its sources
Numbered links point to the page and section each part of the answer came from, so anyone can check it in one click.
Got questions?
Frequently Asked Questions
Common questions about Bring Your Own Model (BYOK)
OpenAI, Anthropic, Google Gemini, and Amazon Bedrock for the chat model, plus OpenAI-compatible vendors (such as Groq, Mistral AI, or OpenRouter) and self-hosted servers such as Ollama or vLLM. Chunk scoring runs on your account too, unless Kelu scores chunks with its own reranking model — there is no separate reranker key to bring.
Not today. The provider you configure applies to every knowledge base in the workspace. You can pick any model your provider serves, for example Claude Sonnet or GPT-4o.
The call fails and the user sees an error; Kelu does not silently switch to its own models. In helpdesk automation nothing is posted and the ticket is left for your team. Revert to default switches the workspace back to Kelu's models.
The key is encrypted at rest with AES-256-GCM and never shown again. Only the request path uses it, and only for the calls that route through the workspace's credential. It's not visible to other workspaces and not used for training or logging beyond the invocation record.
Kelu is priced for the platform (retrieval, connectors, analytics, surfaces). With BYOK, model usage is billed to your own provider account. Without it, answers run on Kelu's models and inference is included in the plan price.
Yes. BYOK is on every plan, including Free. Workspace owners and admins set it up under Settings → AI Model.
Run the platform, keep the LLM contract
BYOK is on every plan. Talk to us about Enterprise if you also need SSO and a countersigned DPA.
Free to start · No credit card required