大模型服务平台 TokenHub 深度思考

功能说明
深度思考模型支持在生成最终答案前先进行推理，提升复杂任务的准确性和可解释性。
适用场景
复杂代码生成、代码修复、代码重构。
数学推导、逻辑分析、多步骤决策。
复杂信息抽取后再综合归纳。
需要更强稳定性和更少推理失误的任务。
开启/关闭深度思考
通过 thinking 参数控制是否开启思考模式。
开启深度思考："thinking":{"type":"enabled"}
关闭深度思考："thinking":{"type":"disabled"}
支持模型
模型名称
model 参数值
默认值及说明
Hy3 preview
hy3-preview
disabled
HY 2.0 Think
hunyuan-2.0-thinking-20251109
enabled 不支持关闭
DeepSeek-V4-Flash
deepseek-v4-flash
enabled
DeepSeek-V4-Pro
deepseek-v4-pro
enabled
Deepseek-v3.2
deepseek-v3.2
disabled
GLM-5.1
glm-5.1
enabled
GLM-5V-Turbo
glm-5v-turbo
enabled
GLM-5-Turbo
glm-5-turbo
enabled
GLM-5
glm-5
enabled
Kimi-K2.6
kimi-k2.6
enabled
Kimi-K2.5
kimi-k2.5
enabled
MiniMax-M2.7
minimax-m2.7
enabled 不支持关闭
MiniMax-M2.5
minimax-m2.5
enabled 不支持关闭
调用示例：开启深度思考
说明：
请将 YOUR_API_KEY 替换为您创建的 API Key。
cURL
Python
Node.js
Java
Go
curl -X POST 'https://tokenhub.tencentmaas.com/v1/chat/completions' \\
  -H 'Content-Type: application/json' \\
  -H 'Authorization: Bearer YOUR_API_KEY' \\
  -d '{
    "model": "hunyuan-2.0-thinking-20251109",
    "messages": [
      {"role": "user", "content": "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？"}
    ],
    "thinking": {"type": "enabled"},
    "stream": false
  }'
from openai import OpenAI
﻿
client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://tokenhub.tencentmaas.com/v1",
)
﻿
response = client.chat.completions.create(
    model="hunyuan-2.0-thinking-20251109",
    messages=[
        {"role": "user", "content": "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？"},
    ],
    extra_body={"thinking": {"type": "enabled"}},
)
﻿
# OpenAI SDK 不直接声明 reasoning_content 字段，需用 getattr 访问
msg = response.choices[0].message
if hasattr(msg, "reasoning_content"):
    print("思考过程:", getattr(msg, "reasoning_content"))
print("最终回答:", msg.content)
import OpenAI from 'openai';
﻿
const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://tokenhub.tencentmaas.com/v1',
});
﻿
// Node.js SDK：thinking 字段直接展开到顶层
const response = await client.chat.completions.create({
  model: 'hunyuan-2.0-thinking-20251109',
  messages: [
    { role: 'user', content: '小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？' },
  ],
  thinking: { type: 'enabled' },
} as any);
﻿
const msg: any = response.choices[0].message;
if (msg.reasoning_content) console.log('思考过程:', msg.reasoning_content);
console.log('最终回答:', msg.content);
import okhttp3.*;
import com.google.gson.Gson;
import java.util.*;
﻿
public class ThinkingChat {
    public static void main(String[] args) throws Exception {
        Map<String, Object> body = new HashMap<>();
        body.put("model", "hunyuan-2.0-thinking-20251109");
        body.put("messages", List.of(
            Map.of("role", "user", "content", "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？")
        ));
        body.put("thinking", Map.of("type", "enabled"));
﻿
        Request request = new Request.Builder()
            .url("https://tokenhub.tencentmaas.com/v1/chat/completions")
            .header("Authorization", "Bearer YOUR_API_KEY")
            .post(RequestBody.create(new Gson().toJson(body), MediaType.parse("application/json")))
            .build();
﻿
        try (Response response = new OkHttpClient().newCall(request).execute()) {
            // 响应体中 message.reasoning_content 为思考过程，message.content 为最终回答
            System.out.println(response.body().string());
        }
    }
}
package main
﻿
import (
    "bytes"
    "encoding/json"
    "fmt"
    "io"
    "net/http"
)
﻿
func main() {
    body, _ := json.Marshal(map[string]interface{}{
        "model": "hunyuan-2.0-thinking-20251109",
        "messages": []map[string]string{
            {"role": "user", "content": "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？"},
        },
        "thinking": map[string]string{"type": "enabled"},
    })
﻿
    req, _ := http.NewRequest("POST",
        "https://tokenhub.tencentmaas.com/v1/chat/completions",
        bytes.NewBuffer(body))
    req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
    req.Header.Set("Content-Type", "application/json")
﻿
    resp, _ := http.DefaultClient.Do(req)
    defer resp.Body.Close()
﻿
    data, _ := io.ReadAll(resp.Body)
    // 响应体中 message.reasoning_content 为思考过程，message.content 为最终回答
    fmt.Println(string(data))
}
调用示例：关闭深度思考
cURL
Python
Node.js
Java
Go
curl -X POST 'https://tokenhub.tencentmaas.com/v1/chat/completions' \\
  -H 'Content-Type: application/json' \\
  -H 'Authorization: Bearer YOUR_API_KEY' \\
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [
      {"role": "user", "content": "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？"}
    ],
    "thinking": {"type": "disabled"},
    "stream": false
  }'
from openai import OpenAI
﻿
client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://tokenhub.tencentmaas.com/v1",
)
﻿
response = client.chat.completions.create(
    model="deepseek-v4-flash",
    messages=[
        {"role": "user", "content": "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？"},
    ],
    extra_body={"thinking": {"type": "disabled"}},
)
print(response.choices[0].message.content)
import OpenAI from 'openai';
﻿
const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://tokenhub.tencentmaas.com/v1',
});
﻿
const response = await client.chat.completions.create({
  model: 'deepseek-v4-flash',
  messages: [
    { role: 'user', content: '小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？' },
  ],
  thinking: { type: 'disabled' },
} as any);
console.log(response.choices[0].message.content);
import okhttp3.*;
import com.google.gson.Gson;
import java.util.*;
﻿
public class DisableThinking {
    public static void main(String[] args) throws Exception {
        Map<String, Object> body = new HashMap<>();
        body.put("model", "deepseek-v4-flash");
        body.put("messages", List.of(
            Map.of("role", "user", "content", "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？")
        ));
        body.put("thinking", Map.of("type", "disabled"));
﻿
        Request request = new Request.Builder()
            .url("https://tokenhub.tencentmaas.com/v1/chat/completions")
            .header("Authorization", "Bearer YOUR_API_KEY")
            .post(RequestBody.create(new Gson().toJson(body), MediaType.parse("application/json")))
            .build();
﻿
        try (Response response = new OkHttpClient().newCall(request).execute()) {
            System.out.println(response.body().string());
        }
    }
}
package main
﻿
import (
    "bytes"
    "encoding/json"
    "fmt"
    "io"
    "net/http"
)
﻿
func main() {
    body, _ := json.Marshal(map[string]interface{}{
        "model": "deepseek-v4-flash",
        "messages": []map[string]string{
            {"role": "user", "content": "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？"},
        },
        "thinking": map[string]string{"type": "disabled"},
    })
﻿
    req, _ := http.NewRequest("POST",
        "https://tokenhub.tencentmaas.com/v1/chat/completions",
        bytes.NewBuffer(body))
    req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
    req.Header.Set("Content-Type", "application/json")
﻿
    resp, _ := http.DefaultClient.Do(req)
    defer resp.Body.Close()
﻿
    data, _ := io.ReadAll(resp.Body)
    fmt.Println(string(data))
}
推理深度配置
通过 reasoning_effort 参数控制推理深度。该参数用于约束模型投入多少推理强度；推理强度越高，通常回答会更充分，但延迟和 token 消耗也会更高。
reasoning_effort 的值
说明
low
轻量推理，推理步数少，速度快，适合简单任务。
medium
平衡模式，适合大多数日常、逻辑适中的复杂任务。
high
深度推理，推理时间最长，思考最深入，适合高难度数学、编程或复杂逻辑推理任务，但延迟和成本最高。
支持模型
模型名称
model 参数值
说明
Hy3 preview
hy3-preview
默认 low
DeepSeek-V4-Flash
deepseek-v4-flash
默认 high
DeepSeek-V4-Pro
deepseek-v4-pro
默认 high
Deepseek-v3.2
deepseek-v3.2
默认 high
调用示例：推理深度配置
cURL
Python
Node.js
Java
Go
curl -X POST 'https://tokenhub.tencentmaas.com/v1/chat/completions' \\
  -H 'Content-Type: application/json' \\
  -H 'Authorization: Bearer YOUR_API_KEY' \\
  -d '{
    "model": "hy3-preview",
    "messages": [
      {"role": "user", "content": "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？"}
    ],
    "stream": false,
    "temperature": 0.9,
    "reasoning_effort": "high"
  }'
from openai import OpenAI
﻿
client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://tokenhub.tencentmaas.com/v1",
)
﻿
response = client.chat.completions.create(
    model="hy3-preview",
    messages=[
        {"role": "user", "content": "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？"},
    ],
    temperature=0.9,
    extra_body={"reasoning_effort": "high"},
)
﻿
msg = response.choices[0].message
if hasattr(msg, "reasoning_content"):
    print("思考过程:", getattr(msg, "reasoning_content"))
print("最终回答:", msg.content)
import OpenAI from 'openai';
﻿
const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://tokenhub.tencentmaas.com/v1',
});
﻿
const response = await client.chat.completions.create({
  model: 'hy3-preview',
  messages: [
    { role: 'user', content: '小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？' },
  ],
  temperature: 0.9,
  reasoning_effort: 'high',
} as any);
﻿
const msg: any = response.choices[0].message;
if (msg.reasoning_content) console.log('思考过程:', msg.reasoning_content);
console.log('最终回答:', msg.content);
import okhttp3.*;
import com.google.gson.Gson;
import java.util.*;
﻿
public class ReasoningEffortChat {
    public static void main(String[] args) throws Exception {
        Map<String, Object> body = new HashMap<>();
        body.put("model", "hy3-preview");
        body.put("messages", List.of(
            Map.of("role", "user", "content", "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？")
        ));
        body.put("temperature", 0.9);
        body.put("reasoning_effort", "high");
﻿
        Request request = new Request.Builder()
            .url("https://tokenhub.tencentmaas.com/v1/chat/completions")
            .header("Authorization", "Bearer YOUR_API_KEY")
            .post(RequestBody.create(new Gson().toJson(body), MediaType.parse("application/json")))
            .build();
﻿
        try (Response response = new OkHttpClient().newCall(request).execute()) {
            System.out.println(response.body().string());
        }
    }
}
package main
﻿
import (
    "bytes"
    "encoding/json"
    "fmt"
    "io"
    "net/http"
)
﻿
func main() {
    body, _ := json.Marshal(map[string]interface{}{
        "model": "hy3-preview",
        "messages": []map[string]string{
            {"role": "user", "content": "小明有5个苹果，给了小红2个，又买了3个，最后还剩几个？"},
        },
        "temperature":      0.9,
        "reasoning_effort": "high",
    })
﻿
    req, _ := http.NewRequest("POST",
        "https://tokenhub.tencentmaas.com/v1/chat/completions",
        bytes.NewBuffer(body))
    req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
    req.Header.Set("Content-Type", "application/json")
﻿
    resp, _ := http.DefaultClient.Do(req)
    defer resp.Body.Close()
﻿
    data, _ := io.ReadAll(resp.Body)
    fmt.Println(string(data))
}
响应示例
启用思考后，响应中会附带 reasoning_content 思考过程字段：
{
  "id": "c95dc87ecce440678c3bb08f5868fee6",
  "object": "chat.completion",
  "created": 1775146546,
  "model": "hy3-preview",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "答案是 6 个。",
        "reasoning_content": "用户现在需要解决的是小明苹果数量变化的问题，首先得理清楚每一步的变化。首先小明一开始有5个苹果，给了小红2个，那这时候应该减去2，对吧？然后又买了3个，这时候要加上3。所以计算的话就是5减2再加3。先算5-2=3，然后3+3=6。"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 22,
    "completion_tokens": 264,
    "total_tokens": 286
  }
}
思考模式下的工具调用，需在每一轮请求都回填历史 reasoning_content，以获取最佳效果，详情请参见 交错式思考模式（Interleaved Thinking）。
模型名称	model 参数值	默认值及说明
Hy3 preview	`hy3-preview`	`disabled`
HY 2.0 Think	`hunyuan-2.0-thinking-20251109`	`enabled` 不支持关闭
DeepSeek-V4-Flash	`deepseek-v4-flash`	`enabled`
DeepSeek-V4-Pro	`deepseek-v4-pro`	`enabled`
Deepseek-v3.2	`deepseek-v3.2`	`disabled`
GLM-5.1	`glm-5.1`	`enabled`
GLM-5V-Turbo	`glm-5v-turbo`	`enabled`
GLM-5-Turbo	`glm-5-turbo`	`enabled`
GLM-5	`glm-5`	`enabled`
Kimi-K2.6	`kimi-k2.6`	`enabled`
Kimi-K2.5	`kimi-k2.5`	`enabled`
MiniMax-M2.7	`minimax-m2.7`	`enabled` 不支持关闭
MiniMax-M2.5	`minimax-m2.5`	`enabled` 不支持关闭
模型名称	model 参数值	说明
Hy3 preview	`hy3-preview`	默认 `low`
DeepSeek-V4-Flash	`deepseek-v4-flash`	默认 `high`
DeepSeek-V4-Pro	`deepseek-v4-pro`	默认 `high`
Deepseek-v3.2	`deepseek-v3.2`	默认 `high`
`reasoning_effort` 的值	说明
`low`	轻量推理，推理步数少，速度快，适合简单任务。
`medium`	平衡模式，适合大多数日常、逻辑适中的复杂任务。
`high`	深度推理，推理时间最长，思考最深入，适合高难度数学、编程或复杂逻辑推理任务，但延迟和成本最高。
深度思考

本页目录：

功能说明

适用场景

开启/关闭深度思考

支持模型

调用示例：开启深度思考

调用示例：关闭深度思考

推理深度配置

支持模型

调用示例：推理深度配置

响应示例
