Create a smart-model policy
curl --request POST \
--url https://{host}/api/llm-gateway/admin/smart-models \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model_alias": "<string>",
"determiner_model_alias": "<string>",
"slots": [
{
"model_alias": "<string>",
"label": "<string>"
}
],
"description": "<string>",
"enabled": true,
"context_breakpoints": [
123
],
"determiner_input_max_chars": 123,
"default_slot_on_failure": 123
}
'import requests
url = "https://{host}/api/llm-gateway/admin/smart-models"
payload = {
"model_alias": "<string>",
"determiner_model_alias": "<string>",
"slots": [
{
"model_alias": "<string>",
"label": "<string>"
}
],
"description": "<string>",
"enabled": True,
"context_breakpoints": [123],
"determiner_input_max_chars": 123,
"default_slot_on_failure": 123
}
headers = {
"Authorization": "Bearer <token>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
body: JSON.stringify({
model_alias: '<string>',
determiner_model_alias: '<string>',
slots: [{model_alias: '<string>', label: '<string>'}],
description: '<string>',
enabled: true,
context_breakpoints: [123],
determiner_input_max_chars: 123,
default_slot_on_failure: 123
})
};
fetch('https://{host}/api/llm-gateway/admin/smart-models', 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://{host}/api/llm-gateway/admin/smart-models",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "POST",
CURLOPT_POSTFIELDS => json_encode([
'model_alias' => '<string>',
'determiner_model_alias' => '<string>',
'slots' => [
[
'model_alias' => '<string>',
'label' => '<string>'
]
],
'description' => '<string>',
'enabled' => true,
'context_breakpoints' => [
123
],
'determiner_input_max_chars' => 123,
'default_slot_on_failure' => 123
]),
CURLOPT_HTTPHEADER => [
"Authorization: Bearer <token>",
"Content-Type: application/json"
],
]);
$response = curl_exec($curl);
$err = curl_error($curl);
curl_close($curl);
if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}package main
import (
"fmt"
"strings"
"net/http"
"io"
)
func main() {
url := "https://{host}/api/llm-gateway/admin/smart-models"
payload := strings.NewReader("{\n \"model_alias\": \"<string>\",\n \"determiner_model_alias\": \"<string>\",\n \"slots\": [\n {\n \"model_alias\": \"<string>\",\n \"label\": \"<string>\"\n }\n ],\n \"description\": \"<string>\",\n \"enabled\": true,\n \"context_breakpoints\": [\n 123\n ],\n \"determiner_input_max_chars\": 123,\n \"default_slot_on_failure\": 123\n}")
req, _ := http.NewRequest("POST", url, payload)
req.Header.Add("Authorization", "Bearer <token>")
req.Header.Add("Content-Type", "application/json")
res, _ := http.DefaultClient.Do(req)
defer res.Body.Close()
body, _ := io.ReadAll(res.Body)
fmt.Println(string(body))
}HttpResponse<String> response = Unirest.post("https://{host}/api/llm-gateway/admin/smart-models")
.header("Authorization", "Bearer <token>")
.header("Content-Type", "application/json")
.body("{\n \"model_alias\": \"<string>\",\n \"determiner_model_alias\": \"<string>\",\n \"slots\": [\n {\n \"model_alias\": \"<string>\",\n \"label\": \"<string>\"\n }\n ],\n \"description\": \"<string>\",\n \"enabled\": true,\n \"context_breakpoints\": [\n 123\n ],\n \"determiner_input_max_chars\": 123,\n \"default_slot_on_failure\": 123\n}")
.asString();require 'uri'
require 'net/http'
url = URI("https://{host}/api/llm-gateway/admin/smart-models")
http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true
request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n \"model_alias\": \"<string>\",\n \"determiner_model_alias\": \"<string>\",\n \"slots\": [\n {\n \"model_alias\": \"<string>\",\n \"label\": \"<string>\"\n }\n ],\n \"description\": \"<string>\",\n \"enabled\": true,\n \"context_breakpoints\": [\n 123\n ],\n \"determiner_input_max_chars\": 123,\n \"default_slot_on_failure\": 123\n}"
response = http.request(request)
puts response.read_body{
"id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
"org_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
"model_alias": "smart-coder",
"enabled": true,
"determiner_model_alias": "gpt-4o-mini",
"slots": [
{
"model_alias": "<string>",
"label": "<string>"
}
],
"context_breakpoints": [
4000,
32000
],
"determiner_input_max_chars": 2000,
"default_slot_on_failure": 123,
"description": "<string>",
"created_at": "2023-11-07T05:31:56Z",
"updated_at": "2023-11-07T05:31:56Z"
}{
"error": "BadRequest",
"message": "request_timeout_secs must be between 1 and 3600"
}LLM Gateway: Smart Models
Create a smart-model policy
POST
/
admin
/
smart-models
Create a smart-model policy
curl --request POST \
--url https://{host}/api/llm-gateway/admin/smart-models \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model_alias": "<string>",
"determiner_model_alias": "<string>",
"slots": [
{
"model_alias": "<string>",
"label": "<string>"
}
],
"description": "<string>",
"enabled": true,
"context_breakpoints": [
123
],
"determiner_input_max_chars": 123,
"default_slot_on_failure": 123
}
'import requests
url = "https://{host}/api/llm-gateway/admin/smart-models"
payload = {
"model_alias": "<string>",
"determiner_model_alias": "<string>",
"slots": [
{
"model_alias": "<string>",
"label": "<string>"
}
],
"description": "<string>",
"enabled": True,
"context_breakpoints": [123],
"determiner_input_max_chars": 123,
"default_slot_on_failure": 123
}
headers = {
"Authorization": "Bearer <token>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
body: JSON.stringify({
model_alias: '<string>',
determiner_model_alias: '<string>',
slots: [{model_alias: '<string>', label: '<string>'}],
description: '<string>',
enabled: true,
context_breakpoints: [123],
determiner_input_max_chars: 123,
default_slot_on_failure: 123
})
};
fetch('https://{host}/api/llm-gateway/admin/smart-models', 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://{host}/api/llm-gateway/admin/smart-models",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "POST",
CURLOPT_POSTFIELDS => json_encode([
'model_alias' => '<string>',
'determiner_model_alias' => '<string>',
'slots' => [
[
'model_alias' => '<string>',
'label' => '<string>'
]
],
'description' => '<string>',
'enabled' => true,
'context_breakpoints' => [
123
],
'determiner_input_max_chars' => 123,
'default_slot_on_failure' => 123
]),
CURLOPT_HTTPHEADER => [
"Authorization: Bearer <token>",
"Content-Type: application/json"
],
]);
$response = curl_exec($curl);
$err = curl_error($curl);
curl_close($curl);
if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}package main
import (
"fmt"
"strings"
"net/http"
"io"
)
func main() {
url := "https://{host}/api/llm-gateway/admin/smart-models"
payload := strings.NewReader("{\n \"model_alias\": \"<string>\",\n \"determiner_model_alias\": \"<string>\",\n \"slots\": [\n {\n \"model_alias\": \"<string>\",\n \"label\": \"<string>\"\n }\n ],\n \"description\": \"<string>\",\n \"enabled\": true,\n \"context_breakpoints\": [\n 123\n ],\n \"determiner_input_max_chars\": 123,\n \"default_slot_on_failure\": 123\n}")
req, _ := http.NewRequest("POST", url, payload)
req.Header.Add("Authorization", "Bearer <token>")
req.Header.Add("Content-Type", "application/json")
res, _ := http.DefaultClient.Do(req)
defer res.Body.Close()
body, _ := io.ReadAll(res.Body)
fmt.Println(string(body))
}HttpResponse<String> response = Unirest.post("https://{host}/api/llm-gateway/admin/smart-models")
.header("Authorization", "Bearer <token>")
.header("Content-Type", "application/json")
.body("{\n \"model_alias\": \"<string>\",\n \"determiner_model_alias\": \"<string>\",\n \"slots\": [\n {\n \"model_alias\": \"<string>\",\n \"label\": \"<string>\"\n }\n ],\n \"description\": \"<string>\",\n \"enabled\": true,\n \"context_breakpoints\": [\n 123\n ],\n \"determiner_input_max_chars\": 123,\n \"default_slot_on_failure\": 123\n}")
.asString();require 'uri'
require 'net/http'
url = URI("https://{host}/api/llm-gateway/admin/smart-models")
http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true
request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n \"model_alias\": \"<string>\",\n \"determiner_model_alias\": \"<string>\",\n \"slots\": [\n {\n \"model_alias\": \"<string>\",\n \"label\": \"<string>\"\n }\n ],\n \"description\": \"<string>\",\n \"enabled\": true,\n \"context_breakpoints\": [\n 123\n ],\n \"determiner_input_max_chars\": 123,\n \"default_slot_on_failure\": 123\n}"
response = http.request(request)
puts response.read_body{
"id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
"org_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
"model_alias": "smart-coder",
"enabled": true,
"determiner_model_alias": "gpt-4o-mini",
"slots": [
{
"model_alias": "<string>",
"label": "<string>"
}
],
"context_breakpoints": [
4000,
32000
],
"determiner_input_max_chars": 2000,
"default_slot_on_failure": 123,
"description": "<string>",
"created_at": "2023-11-07T05:31:56Z",
"updated_at": "2023-11-07T05:31:56Z"
}{
"error": "BadRequest",
"message": "request_timeout_secs must be between 1 and 3600"
}Authorizations
JWT obtained through Barndoor's authentication flow. Pass the token
verbatim in Authorization: Bearer <token>. Use the Barndoor SDK's
loginInteractive() helper to obtain a token in scripts and notebooks.
Body
application/json
Response
The newly created policy
Picks one of several model aliases per request based on a determiner-model classification (e.g. send short prompts to a small model, long prompts to a larger one).
The new alias clients call (must not collide with existing routes)
Example:
"smart-coder"
The model the gateway uses to classify each request
Example:
"gpt-4o-mini"
Minimum array length:
1Show child attributes
Show child attributes
Token counts at which the determiner switches between slots
Example:
[4000, 32000]
Example:
2000
Slot index used when the determiner fails or times out
⌘I