LLM

Decibyl platform supports OpenAI, Google AI Studio, Google Vertex AI, Azure OpenAI, AWS Bedrock, Groq, OpenRouter, Hugging Face, MiniMax, Sarvam, and Decibyl-managed LLMs. There are some models provided by default for you to choose from the drop down.

Caution

LLM providers receive the data needed to generate responses, such as prompts, conversation history, model settings, and any configured tool/function definitions or tool call context. Review the provider’s data processing, retention, model training, and regional hosting policies before using sensitive data.

Decibyl also supports any OpenAI-compatible endpoint, so a model you host yourself — Ollama, vLLM or similar — can be used in place of a provider.

Choosing Decibyl as the provider makes this slot managed — no key to add, billed through your balance. Anything else is your own key. The two are chosen per slot, so your own language model alongside our managed voice is a normal setup; see Model Configurations.

If you don’t find a model in the drop down, you can always add a model manually.