@Shubham Bajaj @henry [Vapi Team] I am using Grog Mistral 7X8 Model and in Logs I see this Does this mean the model is no longer available? https://cdn.discordapp.com/attachments/1354175242087432343/1354175243278745640/image.png?ex=67e45579&is=67e303f9&hm=5f311bcaa14c134b445443d838a62f7b984af1e3b0602ea334851308e331c780&

Hi there! 👋 Yes, I can confirm that the Groq Mistral 7X8 Model (

mistral-8x7b-32768

) has been decommissioned and is no longer supported, as shown in your error message. Based on the information I found, here are some alternative models available in the Vapi ecosystem that you could consider: -

gemma2-9b-it

-

llama-3.1-8b-instant

-

llama3-70b-8192

-

llama3-8b-8192

-

deepseek-r1-distill-llama-70b

-

llama-3.3-70b-versatile

-

llama-3.1-405b-reasoning

Additionally, Vapi supports using any OpenAI-compatible endpoint as the LLM, including services like [OpenRouter, AnyScale, Together AI, or your own server](https://docs.vapi.ai/customization/custom-llm/fine-tuned-openai-models). You can also use open-source LLMs like Mixtral. For more specific recommendations on which model to use instead of the decommissioned Groq Mistral model, I'd recommend checking the URL mentioned in your error message: https://console.grog.com/docs/deprecations. Is there anything else you'd like to know about the available models or how to switch to an alternative? Sources: - [Discord conversation about model decommissioning](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1336171779772907692) - [Discord conversation about available models](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1334534504324403230) - [Vapi documentation on custom LLMs](https://docs.vapi.ai/customization/custom-llm/fine-tuned-openai-models)

Yes PARTH, you are right, the model has been decommissioned, so you have to use an another model.

@Shubham Bajaj How does pricing of openrouter works

You can use your own OpenRouter key so you don't have to pay VAPI for LLM charges. But do note that you might hit a latency using your own key.