https://retellai.com/?utm_source=discord&utm_medium=seo logo
Is Retell planning on supporting open source model...
# 💬・general
s
Is Retell planning on supporting open source models such as Llama 3.3 70b and Qwen3? I've had great results testing these models for conversational AI use cases. They're great at following instructions while being relatively cheap and low latency. Importantly, open source models offer an escape hatch when OpenAI/Anthropic/Google decide to deprecate popular LLMs (as is the case for GPT 4.1, which being retired around October this year). I know Retell supports "custom LLMs", but that requires us to spin up an entire websocket server to manage prompts and LLM interactions. It would be great if I could just give Retell an OpenAI compatible endpoint+API key and still use the agent configuration tools that the Retell platform provides.