List Fallback LLMs

GET

llms

fallback

List Fallback LLMs

curl --request GET \
  --url https://api.dialnexa.com/v1/llms/fallback \
  --header 'Authorization: Bearer <token>'

const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://api.dialnexa.com/v1/llms/fallback', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.dialnexa.com/v1/llms/fallback",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.dialnexa.com/v1/llms/fallback"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("Authorization", "Bearer <token>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

require 'uri'
require 'net/http'

url = URI("https://api.dialnexa.com/v1/llms/fallback")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)
request["Authorization"] = 'Bearer <token>'

response = http.request(request)
puts response.read_body

[
  {
    "id": "llm_mh2k7w9q4x1zpt",
    "name": "Gemini 3.1 Flash Lite",
    "provider": "gemini"
  }
]

{
  "message": "Request validation failed: Validation failed (numeric string is expected)"
}

{
  "message": "API key is missing or invalid. You can generate this by logging in into dashboard"
}

{
  "message": "An unexpected error occurred. Please try again later."
}

Read only Lists the language models that can serve as an agent’s fallback LLM. A fallback LLM takes over mid-call when the primary model is slow or unavailable, so the conversation keeps going instead of stalling.

When to use this

Configuring fallbacks: pick a model from this list, then enable it on the agent with fallback_llm_enabled, llm_fallback_model, and llm_fallback_delay_ms via Create Agent or Update Agent.
Validating configuration: confirm a previously configured fallback model is still offered before publishing an agent change.

The fallback set is maintained separately from the main List LLMs catalog: not every primary model is eligible as a fallback. Choose a fallback from a different provider than the primary model so a provider-wide outage does not take out both. The API-key response contains id, name, and provider; it does not expose supports_tool_calling. Check tool compatibility in the dashboard or MCP catalog. The API validates the effective fallback model when the agent configuration is saved.

List Fallback LLMs results

The response is one page of models in a plain JSON array. The default page size is 20, and requests above 100 models per page are capped at 100. Increment page until the array is shorter than your effective page size. Each entry includes the model’s id and display metadata; use the response schema on this page for the exact fields. See API errors for the standard error format.

Query parameters

Parameter	Description
`page`	Page number, starting at 1. Must be a whole number; `0` or below is treated as `1`.
`limit`	Models per page. Must be a whole number. Default 20; `0` also returns 20, values above 100 are treated as 100, and a negative value fails with `500`.
`sortBy`	Column to sort by, for example `id`, `name` or `provider`. Default id. An unknown column returns `500`.
`sortOrder`	Sort direction. Default ASC. One of `ASC`, `DESC`.

Authorizations

Authorization

string

header

required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Query Parameters

page

number

Page number, starting at 1. Must be a whole number; 0 or below is treated as 1.

Example:

1

limit

number

Models per page. Must be a whole number. Default 20; 0 also returns 20, values above 100 are treated as 100, and a negative value fails with 500.

Example:

20

sortBy

string

Column to sort by, for example id, name or provider. Default id. An unknown column returns 500.

Example:

"id"

sortOrder

enum<string>

Sort direction. Default ASC.

Available options:

ASC,

DESC

Response

Fallback LLM returned successfully.

string

required

Signed LLM ID

Example:

"llm_mfgt5q2x8k4w7r"

name

string

provider

enum<string>

Available options:

google,

openai,

deepseek,

groq,

gemini,

sarvam

List Fallback LLMs

curl --request GET \
  --url https://api.dialnexa.com/v1/llms/fallback \
  --header 'Authorization: Bearer <token>'

const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://api.dialnexa.com/v1/llms/fallback', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.dialnexa.com/v1/llms/fallback",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.dialnexa.com/v1/llms/fallback"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("Authorization", "Bearer <token>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

require 'uri'
require 'net/http'

url = URI("https://api.dialnexa.com/v1/llms/fallback")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)
request["Authorization"] = 'Bearer <token>'

response = http.request(request)
puts response.read_body

[
  {
    "id": "llm_mh2k7w9q4x1zpt",
    "name": "Gemini 3.1 Flash Lite",
    "provider": "gemini"
  }
]

{
  "message": "Request validation failed: Validation failed (numeric string is expected)"
}

{
  "message": "API key is missing or invalid. You can generate this by logging in into dashboard"
}

{
  "message": "An unexpected error occurred. Please try again later."
}

Overview

Endpoints

When to use this

List Fallback LLMs results

Query parameters

Authorizations

Query Parameters

Response

​When to use this

​List Fallback LLMs results

​Query parameters

Authorizations

Query Parameters

Response

When to use this

List Fallback LLMs results

Query parameters