Claude LLM 实时响应

curl --request POST \
  --url https://api.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "user_prompt": "<string>",
  "model_name": "<string>",
  "max_output_tokens": 123,
  "temperature": 123,
  "top_p": 123,
  "web_search": true,
  "force_web_search": true,
  "web_search_country_iso_code": "<string>",
  "web_search_city": "<string>",
  "system_message": "<string>",
  "message_chain": [
    "<string>"
  ],
  "use_reasoning": true,
  "tag": "<string>"
}
'

import requests

url = "https://api.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live"

payload = {
    "user_prompt": "<string>",
    "model_name": "<string>",
    "max_output_tokens": 123,
    "temperature": 123,
    "top_p": 123,
    "web_search": True,
    "force_web_search": True,
    "web_search_country_iso_code": "<string>",
    "web_search_city": "<string>",
    "system_message": "<string>",
    "message_chain": ["<string>"],
    "use_reasoning": True,
    "tag": "<string>"
}
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({
    user_prompt: '<string>',
    model_name: '<string>',
    max_output_tokens: 123,
    temperature: 123,
    top_p: 123,
    web_search: true,
    force_web_search: true,
    web_search_country_iso_code: '<string>',
    web_search_city: '<string>',
    system_message: '<string>',
    message_chain: ['<string>'],
    use_reasoning: true,
    tag: '<string>'
  })
};

fetch('https://api.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live', 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.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live",
  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([
    'user_prompt' => '<string>',
    'model_name' => '<string>',
    'max_output_tokens' => 123,
    'temperature' => 123,
    'top_p' => 123,
    'web_search' => true,
    'force_web_search' => true,
    'web_search_country_iso_code' => '<string>',
    'web_search_city' => '<string>',
    'system_message' => '<string>',
    'message_chain' => [
        '<string>'
    ],
    'use_reasoning' => true,
    'tag' => '<string>'
  ]),
  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.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live"

	payload := strings.NewReader("{\n  \"user_prompt\": \"<string>\",\n  \"model_name\": \"<string>\",\n  \"max_output_tokens\": 123,\n  \"temperature\": 123,\n  \"top_p\": 123,\n  \"web_search\": true,\n  \"force_web_search\": true,\n  \"web_search_country_iso_code\": \"<string>\",\n  \"web_search_city\": \"<string>\",\n  \"system_message\": \"<string>\",\n  \"message_chain\": [\n    \"<string>\"\n  ],\n  \"use_reasoning\": true,\n  \"tag\": \"<string>\"\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.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"user_prompt\": \"<string>\",\n  \"model_name\": \"<string>\",\n  \"max_output_tokens\": 123,\n  \"temperature\": 123,\n  \"top_p\": 123,\n  \"web_search\": true,\n  \"force_web_search\": true,\n  \"web_search_country_iso_code\": \"<string>\",\n  \"web_search_city\": \"<string>\",\n  \"system_message\": \"<string>\",\n  \"message_chain\": [\n    \"<string>\"\n  ],\n  \"use_reasoning\": true,\n  \"tag\": \"<string>\"\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://api.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live")

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  \"user_prompt\": \"<string>\",\n  \"model_name\": \"<string>\",\n  \"max_output_tokens\": 123,\n  \"temperature\": 123,\n  \"top_p\": 123,\n  \"web_search\": true,\n  \"force_web_search\": true,\n  \"web_search_country_iso_code\": \"<string>\",\n  \"web_search_city\": \"<string>\",\n  \"system_message\": \"<string>\",\n  \"message_chain\": [\n    \"<string>\"\n  ],\n  \"use_reasoning\": true,\n  \"tag\": \"<string>\"\n}"

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

{
  "version": "<string>",
  "status_code": 123,
  "status_message": "<string>",
  "time": "<string>",
  "cost": 123,
  "tasks_count": 123,
  "tasks_error": 123,
  "tasks": [
    "<string>"
  ],
  "tasks.id": "<string>",
  "tasks.status_code": 123,
  "tasks.status_message": "<string>",
  "tasks.time": "<string>",
  "tasks.cost": 123,
  "tasks.result_count": 123,
  "tasks.path": [
    "<string>"
  ],
  "tasks.data": {},
  "tasks.result": [
    "<string>"
  ],
  "tasks.result.model_name": "<string>",
  "tasks.result.input_tokens": 123,
  "tasks.result.output_tokens": 123,
  "tasks.result.reasoning_tokens": 123,
  "tasks.result.web_search": true,
  "tasks.result.money_spent": 123,
  "tasks.result.datetime": "<string>",
  "tasks.result.items": [
    "<string>"
  ],
  "tasks.result.items.reasoning": {},
  "tasks.result.items.reasoning.type": "<string>",
  "tasks.result.items.reasoning.sections": [
    "<string>"
  ],
  "tasks.result.items.reasoning.sections.type": "<string>",
  "tasks.result.items.reasoning.sections.text": "<string>",
  "tasks.result.items.message": {},
  "tasks.result.items.message.type": "<string>",
  "tasks.result.items.message.sections": [
    "<string>"
  ],
  "tasks.result.items.message.sections.type": "<string>",
  "tasks.result.items.message.sections.text": "<string>",
  "tasks.result.items.message.sections.annotations": [
    "<string>"
  ],
  "tasks.result.items.message.sections.annotations.title": "<string>",
  "tasks.result.items.message.sections.annotations.url": "<string>",
  "tasks.result.fan_out_queries": [
    "<string>"
  ]
}

POST

https://api.aisa.one/apis/v1

dataforseo

ai_optimization

claude

llm_responses

live

Claude LLM 实时响应

curl --request POST \
  --url https://api.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "user_prompt": "<string>",
  "model_name": "<string>",
  "max_output_tokens": 123,
  "temperature": 123,
  "top_p": 123,
  "web_search": true,
  "force_web_search": true,
  "web_search_country_iso_code": "<string>",
  "web_search_city": "<string>",
  "system_message": "<string>",
  "message_chain": [
    "<string>"
  ],
  "use_reasoning": true,
  "tag": "<string>"
}
'

import requests

url = "https://api.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live"

payload = {
    "user_prompt": "<string>",
    "model_name": "<string>",
    "max_output_tokens": 123,
    "temperature": 123,
    "top_p": 123,
    "web_search": True,
    "force_web_search": True,
    "web_search_country_iso_code": "<string>",
    "web_search_city": "<string>",
    "system_message": "<string>",
    "message_chain": ["<string>"],
    "use_reasoning": True,
    "tag": "<string>"
}
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({
    user_prompt: '<string>',
    model_name: '<string>',
    max_output_tokens: 123,
    temperature: 123,
    top_p: 123,
    web_search: true,
    force_web_search: true,
    web_search_country_iso_code: '<string>',
    web_search_city: '<string>',
    system_message: '<string>',
    message_chain: ['<string>'],
    use_reasoning: true,
    tag: '<string>'
  })
};

fetch('https://api.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live', 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.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live",
  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([
    'user_prompt' => '<string>',
    'model_name' => '<string>',
    'max_output_tokens' => 123,
    'temperature' => 123,
    'top_p' => 123,
    'web_search' => true,
    'force_web_search' => true,
    'web_search_country_iso_code' => '<string>',
    'web_search_city' => '<string>',
    'system_message' => '<string>',
    'message_chain' => [
        '<string>'
    ],
    'use_reasoning' => true,
    'tag' => '<string>'
  ]),
  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.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live"

	payload := strings.NewReader("{\n  \"user_prompt\": \"<string>\",\n  \"model_name\": \"<string>\",\n  \"max_output_tokens\": 123,\n  \"temperature\": 123,\n  \"top_p\": 123,\n  \"web_search\": true,\n  \"force_web_search\": true,\n  \"web_search_country_iso_code\": \"<string>\",\n  \"web_search_city\": \"<string>\",\n  \"system_message\": \"<string>\",\n  \"message_chain\": [\n    \"<string>\"\n  ],\n  \"use_reasoning\": true,\n  \"tag\": \"<string>\"\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.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"user_prompt\": \"<string>\",\n  \"model_name\": \"<string>\",\n  \"max_output_tokens\": 123,\n  \"temperature\": 123,\n  \"top_p\": 123,\n  \"web_search\": true,\n  \"force_web_search\": true,\n  \"web_search_country_iso_code\": \"<string>\",\n  \"web_search_city\": \"<string>\",\n  \"system_message\": \"<string>\",\n  \"message_chain\": [\n    \"<string>\"\n  ],\n  \"use_reasoning\": true,\n  \"tag\": \"<string>\"\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://api.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live")

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  \"user_prompt\": \"<string>\",\n  \"model_name\": \"<string>\",\n  \"max_output_tokens\": 123,\n  \"temperature\": 123,\n  \"top_p\": 123,\n  \"web_search\": true,\n  \"force_web_search\": true,\n  \"web_search_country_iso_code\": \"<string>\",\n  \"web_search_city\": \"<string>\",\n  \"system_message\": \"<string>\",\n  \"message_chain\": [\n    \"<string>\"\n  ],\n  \"use_reasoning\": true,\n  \"tag\": \"<string>\"\n}"

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

{
  "version": "<string>",
  "status_code": 123,
  "status_message": "<string>",
  "time": "<string>",
  "cost": 123,
  "tasks_count": 123,
  "tasks_error": 123,
  "tasks": [
    "<string>"
  ],
  "tasks.id": "<string>",
  "tasks.status_code": 123,
  "tasks.status_message": "<string>",
  "tasks.time": "<string>",
  "tasks.cost": 123,
  "tasks.result_count": 123,
  "tasks.path": [
    "<string>"
  ],
  "tasks.data": {},
  "tasks.result": [
    "<string>"
  ],
  "tasks.result.model_name": "<string>",
  "tasks.result.input_tokens": 123,
  "tasks.result.output_tokens": 123,
  "tasks.result.reasoning_tokens": 123,
  "tasks.result.web_search": true,
  "tasks.result.money_spent": 123,
  "tasks.result.datetime": "<string>",
  "tasks.result.items": [
    "<string>"
  ],
  "tasks.result.items.reasoning": {},
  "tasks.result.items.reasoning.type": "<string>",
  "tasks.result.items.reasoning.sections": [
    "<string>"
  ],
  "tasks.result.items.reasoning.sections.type": "<string>",
  "tasks.result.items.reasoning.sections.text": "<string>",
  "tasks.result.items.message": {},
  "tasks.result.items.message.type": "<string>",
  "tasks.result.items.message.sections": [
    "<string>"
  ],
  "tasks.result.items.message.sections.type": "<string>",
  "tasks.result.items.message.sections.text": "<string>",
  "tasks.result.items.message.sections.annotations": [
    "<string>"
  ],
  "tasks.result.items.message.sections.annotations.title": "<string>",
  "tasks.result.items.message.sections.annotations.url": "<string>",
  "tasks.result.fan_out_queries": [
    "<string>"
  ]
}

把 user_prompt 发给某个 Claude 模型并同步返回回复。model_name 必须来自 get_dataforseo_ai_claude_llm_responses_models——名字形如 claude-sonnet-5，看起来合理的猜测会被拒。十六个 token 的回复实测

0.000744。返回 `model_name`、`input_tokens`、`output_tokens`、`reasoning_tokens`、`web_search`、`money_spent`、`datetime`、`items` 和 `fan_out_queries`。💰 **上游按 token 计费，不是按次**：十六个 token 的回复实测

0.0006，但长回复没有上限，而 AIsa 两种情况都收固定 $0.012。看响应里的 money_spent 才知道一个 prompt 实际花了多少。响应包在信封里：数据在 tasks[0].result，成败在 tasks[0].status_code——请求被拒时 HTTP 仍是 200。

示例

curl -X POST "https://api.aisa.one/apis/v1/dataforseo/ai_optimization/claude/llm_responses/live" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '[{"user_prompt": "...", "model_name": "...", "max_output_tokens": "..."}]'

授权

Authorization

string

header

必填

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

请求体

application/json

user_prompt

string

必填

AI 模型的提示词，必填字段；您希望发送给 AI 模型的问题或任务；可在 user_prompt 字段中指定最多 500 个字符

model_name

string

必填

AI 模型名称，必填字段；model_name 由实际模型名称和版本名称组成；如果指定基础模型名称，默认使用其最新版本；例如，如果指定 claude-opus-4-0，将自动把 claude-opus-4-20250514 设置为 model_name；你可以单独请求 https://api.dataforseo.com/v3/ai_optimization/claude/llm_responses/models，以获取可用 LLM 模型的列表

max_output_tokens

integer

AI 响应中的最大词元数；可选字段；最小值：1；最大值：4096；默认值：2048；注意：如果 web_search 设置为 true，或请求中指定了推理模型，则输出词元数可能超过指定的 max_output_tokens 限制；注意 #2：如果 use_reasoning 设置为 true，max_output_tokens 的最小值为 1025

temperature

number

AI 响应的随机性可选字段较高的值使输出更具多样性；较低的值使输出更聚焦；最小值：0 最大值：1 默认值：0.7注意：temperature 不能与 top_p 在同一请求中同时使用

top_p

number

AI 响应的多样性，可选字段，通过限制词元选择来控制响应的多样性；最小值：0，最大值：1，默认值：null。注意：top_p 不能与 temperature 在同一请求中同时使用

web_search

boolean

启用网络搜索以获取当前信息；可选字段。启用后，AI 模型可以访问并引用当前网络信息；注意：有关支持 web_search 的模型列表，请参阅 Models 端点；默认值：false；该参数的费用可在 Pricing 页面计算

force_web_search

boolean

强制 AI 智能体使用网页搜索；可选字段；要启用此参数，还必须启用 web_search；启用后，将强制 AI 模型访问并引用当前的网页信息；默认值：false；注意：即使将该参数设置为 true，也不能保证响应中会引用网页来源

web_search_country_iso_code

string

位置的 ISO 国家/地区代码，可选字段；可能的值：'AR','AT','AU','BE','BR','CA','CH','CL','CN','DE','DK','ES','FI','FR','GB','HK','ID','IN','IT','JP','KR','MX','MY','NL','NO','NZ','PH','PL','PT','RU','SA','SE','TR','TW','US','ZA'

web_search_city

string

位置的城市名称，可选字段。注意：需指定 web_search_country_iso_code 才能使用此参数

system_message

string

AI 行为指令，可选字段，用于定义 AI 的角色、语气或特定行为；可在 system_message 字段中指定最多 500 个字符

message_chain

string[]

对话历史，可选字段。表示之前对话轮次的消息对象数组；每个对象必须包含 role 和 message 参数：role 字符串，值为 user 或 ai 角色；message 字符串，包含消息内容（最多 500 个字符）；数组中最多可以指定 10 个消息对象；示例："message_chain": [{"role":"user","message":"Hello, what’s up?"},{"role":"ai","message":"Hello! I’m doing well, thank you. How can I assist you today?"}]

use_reasoning

boolean

为 AI 模型启用推理。可选字段。启用后，模型将在生成响应前进行推理。有关支持推理的模型列表，请参阅 Models 端点。默认值：false。注意：如果设置为 true，max_output_tokens 的最小值为 1025。注意 #2：如果设置为 true，force_web_search 必须设置为 false。注意 #3：如果设置为 true，则不能使用 temperature 和 top_p

tag

string

用户定义的任务标识符。可选字段。字符数限制为 255。您可以使用此参数标识任务并将其与结果匹配；您将在响应的 data 对象中找到指定的 tag 值

响应

成功响应

version

string

API 的当前版本

status_code

integer

通用状态码；您可以在此处查看完整的响应码列表。注意：我们强烈建议设计必要的系统来处理相关异常或错误情况

status_message

string

常规信息消息；可在此处查看常规信息消息的完整列表

time

string

执行时间（秒）

cost

number

任务总成本（美元）

tasks_count

integer

tasks 数组中的任务数量

tasks_error

integer

返回错误的 tasks 数组中的任务数量

tasks

string[]

任务数组

tasks.id

string

任务标识符，系统中采用 UUID 格式的唯一任务标识符

tasks.status_code

integer

DataForSEO 生成的任务状态码；范围为：10000-60000；完整响应代码列表可在此处查看

tasks.status_message

string

任务的信息性消息，你可以在此处查看通用信息性消息的完整列表

tasks.time

string

执行时间（秒）

tasks.cost

number

任务成本（美元），包括基础任务价格与 money_spent 值之和

tasks.result_count

integer

结果数组中的元素数量

tasks.path

string[]

URL 路径

tasks.data

object

包含您在 POST 请求中指定的相同参数

tasks.result

string[]

结果数组

tasks.result.model_name

string

所用 AI 模型的名称

tasks.result.input_tokens

integer

输入中的词元数量，即已处理的词元总数

tasks.result.output_tokens

integer

输出中的词元数量，即 AI 响应生成的词元总数

tasks.result.reasoning_tokens

integer

推理词元数用于生成推理内容的词元总数

tasks.result.web_search

boolean

表示是否使用了网络搜索

tasks.result.money_spent

number

AI 词元成本（美元），即第三方 AI 模型提供商根据其定价收取的费用

tasks.result.datetime

string

以 UTC 格式接收结果的日期和时间：“yyyy-mm-dd hh-mm-ss +00:00”，示例：2019-11-15 12:57:46 +00:00

tasks.result.items

string[]

响应条目数组，包含结构化 AI 响应数据

tasks.result.items.reasoning

object

响应中的元素

tasks.result.items.reasoning.type

string

元素类型 = 'reasoning' 注意：此元素仅在推理模型中受支持，且不保证返回

tasks.result.items.reasoning.sections

string[]

推理链分段对象数组，包含 LLM 生成的推理链分段

tasks.result.items.reasoning.sections.type

string

element='summary_text' 的类型

tasks.result.items.reasoning.sections.text

string

推理链部分的文本，用于总结模型的思考过程

tasks.result.items.message

object

响应中的元素

tasks.result.items.message.type

string

元素类型 = 'message'

tasks.result.items.message.sections

string[]

内容分区数组，包含 AI 响应的不同部分

tasks.result.items.message.sections.type

string

元素类型='text'

tasks.result.items.message.sections.text

string

AI 生成的文本内容

tasks.result.items.message.sections.annotations

string[]

用于生成响应的引用数组；如果 web_search 参数未设置为 true，则为 null。注意：即使 web_search 为 true，annotations 也可能返回空值，因为 AI 会尝试检索 Web 信息，但可能找不到相关结果。

tasks.result.items.message.sections.annotations.title

string

引用来源的域名或标题

tasks.result.items.message.sections.annotations.url

string

引用来源的 URL

tasks.result.fan_out_queries

string[]

扩展查询数组，包含从主查询派生的相关搜索查询，以提供更全面的响应

概览

聊天 API

Embeddings 与重排序

搜索 API

Perplexity API

金融 API

Twitter API

Reddit API

Instagram API

Pinterest API

学术搜索 API

WaveInflu API

预测市场 API

CoinGecko

DataForSEO

AgentMail

Apollo

基础功能

Claude LLM 实时响应

示例

授权

请求体

响应

​示例

授权

请求体

响应

示例