Sora Chat 创建 - QuickRouter API 中转接口
▼
Authorization
在 Header 添加参数 Authorization,其值为 Bearer 之后拼接 Token
示例:
Authorization: Bearer ********************
给定一个提示,该模型将返回一个或多个预测的完成,并且还可以返回每个位置的替代标记的概率。 为提供的提示和参数创建完成 官方文档:https://platform.openai.com/docs/api-reference/chat/create
请求参数
Header 参数
Content-Type
string
必需
示例: application/json
Accept
string
必需
示例: application/json
Authorization
string
可选
示例: Bearer {{YOUR_API_KEY}} X-Forwarded-Host
string
可选
示例: api.quickrouter.ai
Body 参数 application/json
model
string
必需
要使用的模型的 ID。有关哪些模型可与聊天 API 一起使用的详细信息,请参阅模型端点兼容性表。
messages
array [object]
必需
至今为止对话所包含的消息列表。Python 代码示例。
role
string
可选
content
string
可选
temperature
integer
可选
使用什么采样温度,介于 0 和 2 之间。较高的值(如 0.8)将使输出更加随机,而较低的值(如 0.2)将使输出更加集中和确定。 我们通常建议改变这个或top_p但不是两者。
top_p
integer
可选
一种替代温度采样的方法,称为核采样,其中模型考虑具有 top_p 概率质量的标记的结果。所以 0.1 意味着只考虑构成前 10% 概率质量的标记。 我们通常建议改变这个或temperature但不是两者。
n
integer
可选
默认为 1 为每个输入消息生成多少个聊天补全选择。
stream
boolean
可选
默认为 false 如果设置,则像在 ChatGPT 中一样会发送部分消息增量。标记将以仅数据的服务器发送事件的形式发送,这些事件在可用时,并在 data: [DONE] 消息终止流。Python 代码示例。
stop
string
可选
默认为 null 最多 4 个序列,API 将停止进一步生成标记。
max_tokens
integer
可选
默认为 inf 在聊天补全中生成的最大标记数。 输入标记和生成标记的总长度受模型的上下文长度限制。计算标记的 Python 代码示例。
presence_penalty
number
可选
-2.0 和 2.0 之间的数字。正值会根据到目前为止是否出现在文本中来惩罚新标记,从而增加模型谈论新主题的可能性。 查看有关频率和存在惩罚的更多信息。
frequency_penalty
number
可选
默认为 0 -2.0 到 2.0 之间的数字。正值根据文本目前的存在频率惩罚新标记,降低模型重复相同行的可能性。 有关频率和存在惩罚的更多信息。
logit_bias
null
可选
修改指定标记出现在补全中的可能性。 接受一个 JSON 对象,该对象将标记(由标记器指定的标记 ID)映射到相关的偏差值(-100 到 100)。从数学上讲,偏差在对模型进行采样之前添加到模型生成的 logit 中。确切效果因模型而异,但-1 和 1 之间的值应减少或增加相关标记的选择可能性;如-100 或 100 这样的值应导致相关标记的禁用或独占选择。
user
string
可选
代表您的最终用户的唯一标识符,可以帮助 OpenAI 监控和检测滥用行为。了解更多。
response_format
object
可选
指定模型必须输出的格式的对象。 将 { "type": "json_object" } 启用 JSON 模式,这可以确保模型生成的消息是有效的 JSON。 重要提示:使用 JSON 模式时,还必须通过系统或用户消息指示模型生成 JSON。如果不这样做,模型可能会生成无休止的空白流,直到生成达到令牌限制,从而导致延迟增加和请求“卡住”的外观。另请注意,如果 finish_reason="length",则消息内容可能会被部分切断,这表示生成超过了 max_tokens 或对话超过了最大上下文长度。 显示属性
seen
integer
可选
此功能处于测试阶段。如果指定,我们的系统将尽最大努力确定性地进行采样,以便使用相同的种子和参数进行重复请求应返回相同的结果。不能保证确定性,您应该参考 system_fingerprint 响应参数来监控后端的更改。
tools
array[string]
必需
模型可以调用的一组工具列表。目前,只支持作为工具的函数。使用此功能来提供模型可以为之生成 JSON 输入的函数列表。
tool_choice
object
必需
控制模型调用哪个函数(如果有的话)。none 表示模型不会调用函数,而是生成消息。auto 表示模型可以在生成消息和调用函数之间进行选择。通过 {"type": "function", "function": {"name": "my_function"}} 强制模型调用该函数。 如果没有函数存在,默认为 none。如果有函数存在,默认为 auto。 显示可能的类型 { "model": "sora-2", "max_tokens": 1000, "messages": [ { "role": "user", "content": "an astronaut golden retriever named Sora levitates around an intergalactic pup-themed space station with a tiny jet back that propels him. gorgeous specular lighting and comets fly through the sky, retro-future astro-themed music plays in the background. light glimmers off the dog's eyes. the dog initially propels towards the space station with the doors opening to let him in. the shot then changes. now inside the space station, many tennis balls are flying around in zero gravity. the dog's astronaut helmet opens up so he can grab one. 35mm film, the intricate details and texturing of the dog's hair are clearly visible and the light of the comets shimmers off the fur." } ], "stream": true } 请求
示例
{
"model": "sora-2",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "an astronaut golden retriever named Sora levitates around an intergalactic pup-themed space station with a tiny jet back that propels him. gorgeous specular lighting and comets fly through the sky, retro-future astro-themed music plays in the background. light glimmers off the dog's eyes. the dog initially propels towards the space station with the doors opening to let him in. the shot then changes. now inside the space station, many tennis balls are flying around in zero gravity. the dog's astronaut helmet opens up so he can grab one. 35mm film, the intricate details and texturing of the dog's hair are clearly visible and the light of the comets shimmers off the fur."
}
],
"stream": true
}
请求示例代码
curl --location --request POST 'https://api.quickrouter.ai/v1/chat/completions' \
--header 'Accept: application/json' \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data-raw '{
"model": "sora-2",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "an astronaut golden retriever named Sora levitates around an intergalactic pup-themed space station with a tiny jet back that propels him. gorgeous specular lighting and comets fly through the sky, retro-future astro-themed music plays in the background. light glimmers off the dog's eyes. the dog initially propels towards the space station with the doors opening to let him in. the shot then changes. now inside the space station, many tennis balls are flying around in zero gravity. the dog's astronaut helmet opens up so he can grab one. 35mm film, the intricate details and texturing of the dog's hair are clearly visible and the light of the comets shimmers off the fur."
}
],
"stream": true
}'
var myHeaders = new Headers();
myHeaders.append("Accept", "application/json");
myHeaders.append("Authorization", "Bearer YOUR_API_KEY");
myHeaders.append("Content-Type", "application/json");
var raw = JSON.stringify({
"model": "sora-2",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "an astronaut golden retriever named Sora levitates around an intergalactic pup-themed space station with a tiny jet back that propels him. gorgeous specular lighting and comets fly through the sky, retro-future astro-themed music plays in the background. light glimmers off the dog's eyes. the dog initially propels towards the space station with the doors opening to let him in. the shot then changes. now inside the space station, many tennis balls are flying around in zero gravity. the dog's astronaut helmet opens up so he can grab one. 35mm film, the intricate details and texturing of the dog's hair are clearly visible and the light of the comets shimmers off the fur."
}
],
"stream": true
});
var requestOptions = {
method: 'POST',
headers: myHeaders,
body: raw,
redirect: 'follow'
};
fetch("https://api.quickrouter.ai/v1/chat/completions", requestOptions)
.then(response => response.text())
.then(result => console.log(result))
.catch(error => console.log('error', error));
import http.client
import json
conn = http.client.HTTPSConnection("api.quickrouter.ai")
payload = json.dumps({
"model": "sora-2",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "an astronaut golden retriever named Sora levitates around an intergalactic pup-themed space station with a tiny jet back that propels him. gorgeous specular lighting and comets fly through the sky, retro-future astro-themed music plays in the background. light glimmers off the dog's eyes. the dog initially propels towards the space station with the doors opening to let him in. the shot then changes. now inside the space station, many tennis balls are flying around in zero gravity. the dog's astronaut helmet opens up so he can grab one. 35mm film, the intricate details and texturing of the dog's hair are clearly visible and the light of the comets shimmers off the fur."
}
],
"stream": True
})
headers = {
'Accept': 'application/json',
'Authorization': 'Bearer YOUR_API_KEY',
'Content-Type': 'application/json'
}
conn.request("POST", "/v1/chat/completions", payload, headers)
res = conn.getresponse()
data = res.read()
print(data.decode("utf-8"))
返回响应
响应参数 🟢 200 OK · application/json
id
string
必需
object
string
必需
created
integer
必需
choices
array [object]
必需
index
integer
可选
message
object
可选
finish_reason
string
可选
usage
object
必需
prompt_tokens
integer
必需
completion_tokens
integer
必需
total_tokens
integer
必需
{ "id": "chatcmpl-123", "object": "chat.completion", "created": 1677652288, "choices": [ { "index": 0, "message": { "role": "assistant", "content": "\n\nHello there, how may I assist you today?" }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 9, "completion_tokens": 12, "total_tokens": 21 } }
示例
{
"id": "chatcmpl-123",
"object": "chat.completion",
"created": 1677652288,
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "\n\nHello there, how may I assist you today?"
},
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 9,
"completion_tokens": 12,
"total_tokens": 21
}
}