Create a chat completion (OpenAI-compatible)
curl --request POST \
--url https://api.impossibl.com/v1/chat/completions \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "openai/gpt-4o-mini",
"messages": [
{
"content": "<string>",
"tool_call_id": "<string>",
"tool_calls": [
{
"id": "<string>",
"type": "function",
"function": {
"name": "<string>",
"arguments": "<string>"
}
}
]
}
],
"stream": false,
"stream_options": {
"include_usage": true
},
"temperature": 123,
"top_p": 123,
"max_tokens": 123,
"max_completion_tokens": 123,
"tools": [
{
"type": "function",
"function": {
"name": "<string>",
"description": "<string>",
"parameters": {}
}
}
],
"parallel_tool_calls": true
}
'import requests
url = "https://api.impossibl.com/v1/chat/completions"
payload = {
"model": "openai/gpt-4o-mini",
"messages": [
{
"content": "<string>",
"tool_call_id": "<string>",
"tool_calls": [
{
"id": "<string>",
"type": "function",
"function": {
"name": "<string>",
"arguments": "<string>"
}
}
]
}
],
"stream": False,
"stream_options": { "include_usage": True },
"temperature": 123,
"top_p": 123,
"max_tokens": 123,
"max_completion_tokens": 123,
"tools": [
{
"type": "function",
"function": {
"name": "<string>",
"description": "<string>",
"parameters": {}
}
}
],
"parallel_tool_calls": True
}
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: 'openai/gpt-4o-mini',
messages: [
{
content: '<string>',
tool_call_id: '<string>',
tool_calls: [
{
id: '<string>',
type: 'function',
function: {name: '<string>', arguments: '<string>'}
}
]
}
],
stream: false,
stream_options: {include_usage: true},
temperature: 123,
top_p: 123,
max_tokens: 123,
max_completion_tokens: 123,
tools: [
{
type: 'function',
function: {name: '<string>', description: '<string>', parameters: {}}
}
],
parallel_tool_calls: true
})
};
fetch('https://api.impossibl.com/v1/chat/completions', 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.impossibl.com/v1/chat/completions",
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' => 'openai/gpt-4o-mini',
'messages' => [
[
'content' => '<string>',
'tool_call_id' => '<string>',
'tool_calls' => [
[
'id' => '<string>',
'type' => 'function',
'function' => [
'name' => '<string>',
'arguments' => '<string>'
]
]
]
]
],
'stream' => false,
'stream_options' => [
'include_usage' => true
],
'temperature' => 123,
'top_p' => 123,
'max_tokens' => 123,
'max_completion_tokens' => 123,
'tools' => [
[
'type' => 'function',
'function' => [
'name' => '<string>',
'description' => '<string>',
'parameters' => [
]
]
]
],
'parallel_tool_calls' => true
]),
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://api.impossibl.com/v1/chat/completions"
payload := strings.NewReader("{\n \"model\": \"openai/gpt-4o-mini\",\n \"messages\": [\n {\n \"content\": \"<string>\",\n \"tool_call_id\": \"<string>\",\n \"tool_calls\": [\n {\n \"id\": \"<string>\",\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"arguments\": \"<string>\"\n }\n }\n ]\n }\n ],\n \"stream\": false,\n \"stream_options\": {\n \"include_usage\": true\n },\n \"temperature\": 123,\n \"top_p\": 123,\n \"max_tokens\": 123,\n \"max_completion_tokens\": 123,\n \"tools\": [\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"description\": \"<string>\",\n \"parameters\": {}\n }\n }\n ],\n \"parallel_tool_calls\": true\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://api.impossibl.com/v1/chat/completions")
.header("Authorization", "Bearer <token>")
.header("Content-Type", "application/json")
.body("{\n \"model\": \"openai/gpt-4o-mini\",\n \"messages\": [\n {\n \"content\": \"<string>\",\n \"tool_call_id\": \"<string>\",\n \"tool_calls\": [\n {\n \"id\": \"<string>\",\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"arguments\": \"<string>\"\n }\n }\n ]\n }\n ],\n \"stream\": false,\n \"stream_options\": {\n \"include_usage\": true\n },\n \"temperature\": 123,\n \"top_p\": 123,\n \"max_tokens\": 123,\n \"max_completion_tokens\": 123,\n \"tools\": [\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"description\": \"<string>\",\n \"parameters\": {}\n }\n }\n ],\n \"parallel_tool_calls\": true\n}")
.asString();require 'uri'
require 'net/http'
url = URI("https://api.impossibl.com/v1/chat/completions")
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\": \"openai/gpt-4o-mini\",\n \"messages\": [\n {\n \"content\": \"<string>\",\n \"tool_call_id\": \"<string>\",\n \"tool_calls\": [\n {\n \"id\": \"<string>\",\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"arguments\": \"<string>\"\n }\n }\n ]\n }\n ],\n \"stream\": false,\n \"stream_options\": {\n \"include_usage\": true\n },\n \"temperature\": 123,\n \"top_p\": 123,\n \"max_tokens\": 123,\n \"max_completion_tokens\": 123,\n \"tools\": [\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"description\": \"<string>\",\n \"parameters\": {}\n }\n }\n ],\n \"parallel_tool_calls\": true\n}"
response = http.request(request)
puts response.read_body{
"id": "chatcmpl-...",
"object": "chat.completion",
"created": 123,
"model": "openai/gpt-4o-mini",
"choices": [
{
"index": 123,
"finish_reason": "stop",
"message": {
"role": "assistant",
"content": "<string>",
"tool_calls": [
{
"id": "<string>",
"type": "function",
"function": {
"name": "<string>",
"arguments": "<string>"
}
}
]
}
}
],
"usage": {
"prompt_tokens": 123,
"completion_tokens": 123,
"total_tokens": 123,
"prompt_tokens_details": {
"cached_tokens": 123
},
"completion_tokens_details": {
"reasoning_tokens": 123
}
},
"service_tier": "<string>"
}{
"error": {
"message": "`model` and `messages` are required",
"type": "invalid_request_error"
}
}{
"error": {
"message": "Invalid API key",
"type": "authentication_error"
}
}{
"error": {
"message": "Insufficient credits. Top up to continue.",
"type": "insufficient_quota"
}
}{
"error": {
"message": "Unknown model: foo/bar",
"type": "invalid_request_error"
}
}{
"error": {
"message": "permission-denied: ...",
"type": "upstream_error"
}
}{
"error": {
"message": "Provider not configured: missing OPENAI_API_KEY",
"type": "provider_not_configured"
}
}API Reference
Create a chat completion (OpenAI-compatible)
Drop-in compatible with OpenAI’s Chat Completions API. Set stream: true for SSE.
POST
/
v1
/
chat
/
completions
Create a chat completion (OpenAI-compatible)
curl --request POST \
--url https://api.impossibl.com/v1/chat/completions \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "openai/gpt-4o-mini",
"messages": [
{
"content": "<string>",
"tool_call_id": "<string>",
"tool_calls": [
{
"id": "<string>",
"type": "function",
"function": {
"name": "<string>",
"arguments": "<string>"
}
}
]
}
],
"stream": false,
"stream_options": {
"include_usage": true
},
"temperature": 123,
"top_p": 123,
"max_tokens": 123,
"max_completion_tokens": 123,
"tools": [
{
"type": "function",
"function": {
"name": "<string>",
"description": "<string>",
"parameters": {}
}
}
],
"parallel_tool_calls": true
}
'import requests
url = "https://api.impossibl.com/v1/chat/completions"
payload = {
"model": "openai/gpt-4o-mini",
"messages": [
{
"content": "<string>",
"tool_call_id": "<string>",
"tool_calls": [
{
"id": "<string>",
"type": "function",
"function": {
"name": "<string>",
"arguments": "<string>"
}
}
]
}
],
"stream": False,
"stream_options": { "include_usage": True },
"temperature": 123,
"top_p": 123,
"max_tokens": 123,
"max_completion_tokens": 123,
"tools": [
{
"type": "function",
"function": {
"name": "<string>",
"description": "<string>",
"parameters": {}
}
}
],
"parallel_tool_calls": True
}
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: 'openai/gpt-4o-mini',
messages: [
{
content: '<string>',
tool_call_id: '<string>',
tool_calls: [
{
id: '<string>',
type: 'function',
function: {name: '<string>', arguments: '<string>'}
}
]
}
],
stream: false,
stream_options: {include_usage: true},
temperature: 123,
top_p: 123,
max_tokens: 123,
max_completion_tokens: 123,
tools: [
{
type: 'function',
function: {name: '<string>', description: '<string>', parameters: {}}
}
],
parallel_tool_calls: true
})
};
fetch('https://api.impossibl.com/v1/chat/completions', 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.impossibl.com/v1/chat/completions",
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' => 'openai/gpt-4o-mini',
'messages' => [
[
'content' => '<string>',
'tool_call_id' => '<string>',
'tool_calls' => [
[
'id' => '<string>',
'type' => 'function',
'function' => [
'name' => '<string>',
'arguments' => '<string>'
]
]
]
]
],
'stream' => false,
'stream_options' => [
'include_usage' => true
],
'temperature' => 123,
'top_p' => 123,
'max_tokens' => 123,
'max_completion_tokens' => 123,
'tools' => [
[
'type' => 'function',
'function' => [
'name' => '<string>',
'description' => '<string>',
'parameters' => [
]
]
]
],
'parallel_tool_calls' => true
]),
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://api.impossibl.com/v1/chat/completions"
payload := strings.NewReader("{\n \"model\": \"openai/gpt-4o-mini\",\n \"messages\": [\n {\n \"content\": \"<string>\",\n \"tool_call_id\": \"<string>\",\n \"tool_calls\": [\n {\n \"id\": \"<string>\",\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"arguments\": \"<string>\"\n }\n }\n ]\n }\n ],\n \"stream\": false,\n \"stream_options\": {\n \"include_usage\": true\n },\n \"temperature\": 123,\n \"top_p\": 123,\n \"max_tokens\": 123,\n \"max_completion_tokens\": 123,\n \"tools\": [\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"description\": \"<string>\",\n \"parameters\": {}\n }\n }\n ],\n \"parallel_tool_calls\": true\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://api.impossibl.com/v1/chat/completions")
.header("Authorization", "Bearer <token>")
.header("Content-Type", "application/json")
.body("{\n \"model\": \"openai/gpt-4o-mini\",\n \"messages\": [\n {\n \"content\": \"<string>\",\n \"tool_call_id\": \"<string>\",\n \"tool_calls\": [\n {\n \"id\": \"<string>\",\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"arguments\": \"<string>\"\n }\n }\n ]\n }\n ],\n \"stream\": false,\n \"stream_options\": {\n \"include_usage\": true\n },\n \"temperature\": 123,\n \"top_p\": 123,\n \"max_tokens\": 123,\n \"max_completion_tokens\": 123,\n \"tools\": [\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"description\": \"<string>\",\n \"parameters\": {}\n }\n }\n ],\n \"parallel_tool_calls\": true\n}")
.asString();require 'uri'
require 'net/http'
url = URI("https://api.impossibl.com/v1/chat/completions")
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\": \"openai/gpt-4o-mini\",\n \"messages\": [\n {\n \"content\": \"<string>\",\n \"tool_call_id\": \"<string>\",\n \"tool_calls\": [\n {\n \"id\": \"<string>\",\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"arguments\": \"<string>\"\n }\n }\n ]\n }\n ],\n \"stream\": false,\n \"stream_options\": {\n \"include_usage\": true\n },\n \"temperature\": 123,\n \"top_p\": 123,\n \"max_tokens\": 123,\n \"max_completion_tokens\": 123,\n \"tools\": [\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"<string>\",\n \"description\": \"<string>\",\n \"parameters\": {}\n }\n }\n ],\n \"parallel_tool_calls\": true\n}"
response = http.request(request)
puts response.read_body{
"id": "chatcmpl-...",
"object": "chat.completion",
"created": 123,
"model": "openai/gpt-4o-mini",
"choices": [
{
"index": 123,
"finish_reason": "stop",
"message": {
"role": "assistant",
"content": "<string>",
"tool_calls": [
{
"id": "<string>",
"type": "function",
"function": {
"name": "<string>",
"arguments": "<string>"
}
}
]
}
}
],
"usage": {
"prompt_tokens": 123,
"completion_tokens": 123,
"total_tokens": 123,
"prompt_tokens_details": {
"cached_tokens": 123
},
"completion_tokens_details": {
"reasoning_tokens": 123
}
},
"service_tier": "<string>"
}{
"error": {
"message": "`model` and `messages` are required",
"type": "invalid_request_error"
}
}{
"error": {
"message": "Invalid API key",
"type": "authentication_error"
}
}{
"error": {
"message": "Insufficient credits. Top up to continue.",
"type": "insufficient_quota"
}
}{
"error": {
"message": "Unknown model: foo/bar",
"type": "invalid_request_error"
}
}{
"error": {
"message": "permission-denied: ...",
"type": "upstream_error"
}
}{
"error": {
"message": "Provider not configured: missing OPENAI_API_KEY",
"type": "provider_not_configured"
}
}Authorizations
Gateway API key (imp-rt-...) from POST /v1/accounts or /v1/keys.
Body
application/json
Example:
"openai/gpt-4o-mini"
Show child attributes
Show child attributes
Show child attributes
Show child attributes
Function tools the model may call (relay mode — the gateway does not execute them).
Show child attributes
Show child attributes
auto, none, required, or { type: function, function: { name } }.
Available options:
auto, none, required Translated to model and route-specific reasoning controls. A level the model lacks moves to the nearest one it has (ties to the lower), so max runs at the model's highest level. A model's reasoning field on GET /v1/models lists the levels that make a difference; the x-impossibl-reasoning response header reports what was applied.
Available options:
none, minimal, low, medium, high, xhigh, max Processing tier. Omitted, null, auto and default are standard processing at the model's normal rates. ultrafast is OpenAI's low-latency tier, available only for openai/gpt-6-astra and openai/gpt-6.1-sol: it is served only by OpenAI's own API (never another host, never a silent fallback to standard) and billed at Ultrafast rates. fast uses native Anthropic only for anthropic/claude-opus-5-5, billed at Fast rates; Foundry pins conflict with Fast. priority and flex return 400. See the Speed tiers guide.
Available options:
auto, default, ultrafast, fast Request-level Anthropic prompt-caching breakpoint. The gateway places one breakpoint on the last cacheable block and advances it as the conversation grows. A per-part cache_control anywhere in the request takes precedence and this is ignored. Non-Anthropic routes cache implicitly and ignore it.
Show child attributes
Show child attributes
Response
A chat.completion object. When stream: true, the body is instead an
SSE stream of chat.completion.chunk objects terminated by data: [DONE].
Non-streaming chat completion (OpenAI chat.completion).
Example:
"chatcmpl-..."
Allowed value:
"chat.completion"Unix seconds
Example:
"openai/gpt-4o-mini"
Show child attributes
Show child attributes
OpenAI-style token usage.
Show child attributes
Show child attributes
Present only when the request set service_tier: the tier the provider reports it actually processed the request at: fast or standard for Anthropic, ultrafast or default for OpenAI. A confirmed standard downgrade is billed at standard rates. null when unreported; missing or unknown metadata bills the requested premium tier. Any reported string is preserved verbatim, including unknown values. The gateway never invents a delivered tier.
⌘I