When are we getting grok 4.20? I was told by a team member grok-4-1-fast-reasoning, and grok-4-1-fast-reasoning were coming soon about 4 months ago, yet here we are and they're not here. They accept the same params (and 4.20 does too) as existing models so it's literally just adding them as options and migrating the db. in the meantime, If i use them using the custom llm workaround am i losing anything? as i understand it, vapi only has fallbacks built in when using openAI right? so if i go custom with a non-OpenAI model, im not missing out on a fallback path, because their's not one anyway?

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards, Vapi Customer Support Team

If you use the custom LLM workaround for Grok models, you aren't missing out on a fallback path—Vapi's built-in fallback is only available for OpenAI models. Here's a quick example of how to set up your custom LLM endpoint with Flask:

python
from flask import Flask, request, jsonify
import openai

app = Flask(__name__)
openai.api_key = "YOUR_OPENAI_API_KEY"

@app.route("/chat/completions", methods=["POST"])
def chat_completions():
    data = request.get_json()
    # ...extract prompt and history...
    response = openai.ChatCompletion.create(
        model="gpt-4.1-mini",
        messages=[{"role": "system", "content": "You are a helpful assistant."}]
    )
    # ...format response for Vapi...
    return jsonify(formatted_response)

if __name__ == "__main__":
    app.run(debug=True, port=5000)

Just point Vapi to your server's public URL—no fallback will be applied for custom LLMs. Reference: See the [Vapi custom LLM integration guide](https://docs.vapi.ai/customization/custom-llm/using-your-server) for more details. Source: - [Connecting Your Custom LLM to Vapi: A Comprehensive Guide](https://docs.vapi.ai/customization/custom-llm/using-your-server)

Using Custom LLM for Grok doesn't lose fallback behavior because xAI/Grok native integration doesn't have fallback support either. Only OpenAI and Vapi providers support

fallbackModels

yeah unfortunately it adds extra latency instead of just letting vapi call the llm directly

Hi, we recommend please try different llm if you want to add fallback models and please let us know if any issues. Thank you