Alibaba Cloud Model Studio:ディープシンキング

サンプルコード

from openai import OpenAI
import os

# OpenAI クライアントを初期化
client = OpenAI(
    # 環境変数を設定していない場合は、次の内容を Model Studio API キーに置き換えます： api_key="sk-xxx"
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    # 以下はシンガポールリージョンの base_url です。バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/compatible-mode/v1 に置き換えてください
    # 北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/compatible-mode/v1 に置き換えてください
    base_url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
)

messages = [{"role": "user", "content": "Who are you"}]

completion = client.chat.completions.create(
    model="qwen-plus",  # 必要に応じて、他のディープシンキングモデルに置き換えることができます
    messages=messages,
    extra_body={"enable_thinking": True},
    stream=True,
    stream_options={
        "include_usage": True
    },
)

reasoning_content = ""  # 完全な思考プロセス
answer_content = ""  # 完全な応答
is_answering = False  # 応答フェーズが開始されたかどうかを示す
print("\n" + "=" * 20 + "Thinking process" + "=" * 20 + "\n")

for chunk in completion:
    if not chunk.choices:
        print("\nUsage:")
        print(chunk.usage)
        continue

    delta = chunk.choices[0].delta

    # 思考コンテンツのみを収集
    if hasattr(delta, "reasoning_content") and delta.reasoning_content is not None:
        if not is_answering:
            print(delta.reasoning_content, end="", flush=True)
        reasoning_content += delta.reasoning_content

    # コンテンツを受信した場合、応答を開始
    if hasattr(delta, "content") and delta.content:
        if not is_answering:
            print("\n" + "=" * 20 + "Full response" + "=" * 20 + "\n")
            is_answering = True
        print(delta.content, end="", flush=True)
        answer_content += delta.content

応答

====================Thinking process====================

Okay, the user asked "Who are you", so I need to provide an accurate and friendly response. First, I need to confirm my identity, which is Qwen, developed by the Tongyi Lab under Alibaba Group. Next, I should explain my main functions, such as answering questions, creating text, and logical reasoning. At the same time, I should maintain a friendly tone and avoid being too technical to make the user feel at ease. I also need to avoid using complex terminology to ensure the response is concise and clear. In addition, I might need to add some interactive elements, inviting the user to ask questions to encourage further communication. Finally, I will check if I have missed any important information, such as my Chinese name "Tongyi Qianwen" and English name "Qwen", along with my parent company and lab. I need to ensure the response is comprehensive and meets the user's expectations.
====================Full response====================

Hello! I am Qwen, an ultra-large language model independently developed by the Tongyi Lab under Alibaba Group. I can answer questions, create text, perform logical reasoning, and write code, with the goal of providing users with high-quality information and services. You can call me Qwen, or just Tongyi Qianwen. How can I help you?

サンプルコード

import OpenAI from "openai";
import process from 'process';

// OpenAI クライアントを初期化
const openai = new OpenAI({
    apiKey: process.env.DASHSCOPE_API_KEY, // 環境変数から読み込み
    // 以下はシンガポールリージョンの baseURL です。バージニアリージョンのモデルを使用する場合は、baseURL を https://dashscope-us.aliyuncs.com/compatible-mode/v1 に置き換えてください 
    // 北京リージョンのモデルを使用する場合は、baseURL を https://dashscope.aliyuncs.com/compatible-mode/v1 に置き換えてください
    baseURL: 'https://dashscope-intl.aliyuncs.com/compatible-mode/v1'
});

let reasoningContent = '';
let answerContent = '';
let isAnswering = false;

async function main() {
    try {
        const messages = [{ role: 'user', content: 'Who are you' }];
        const stream = await openai.chat.completions.create({
            model: 'qwen-plus',
            messages,
            stream: true,
            enable_thinking: true
        });
        console.log('\n' + '='.repeat(20) + 'Thinking process' + '='.repeat(20) + '\n');

        for await (const chunk of stream) {
            if (!chunk.choices?.length) {
                console.log('\nUsage:');
                console.log(chunk.usage);
                continue;
            }

            const delta = chunk.choices[0].delta;
            
            // 思考コンテンツのみを収集
            if (delta.reasoning_content !== undefined && delta.reasoning_content !== null) {
                if (!isAnswering) {
                    process.stdout.write(delta.reasoning_content);
                }
                reasoningContent += delta.reasoning_content;
            }

            // コンテンツを受信した場合、応答を開始
            if (delta.content !== undefined && delta.content) {
                if (!isAnswering) {
                    console.log('\n' + '='.repeat(20) + 'Full response' + '='.repeat(20) + '\n');
                    isAnswering = true;
                }
                process.stdout.write(delta.content);
                answerContent += delta.content;
            }
        }
    } catch (error) {
        console.error('Error:', error);
    }
}

main();

戻り値

====================Thinking process====================

Okay, the user asked "Who are you", so I need to answer about my identity. First, I should clearly state that I am Qwen, an ultra-large language model developed by Alibaba Cloud. Next, I can mention my main functions, such as answering questions, creating text, and logical reasoning. I should also emphasize my multilingual support, including Chinese and English, so the user knows I can handle requests in different languages. In addition, I might need to explain my application scenarios, such as helping with study, work, and daily life. However, the user's question is quite direct, so I probably do not need to provide too much detailed information. I should keep it concise and clear. At the same time, I need to ensure a friendly tone and invite the user to ask further questions. I will check if I have missed any important information, such as my version or latest updates, but the user probably does not need that much detail. Finally, I will confirm that the response is accurate and contains no errors.
====================Full response====================

I am Qwen, an ultra-large language model independently developed by the Tongyi Lab under Alibaba Group. I can perform various tasks such as answering questions, creating text, logical reasoning, and coding. I support multiple languages, including Chinese and English. If you have any questions or need help, feel free to let me know!

サンプルコード

# ======= 重要 =======
# 以下はシンガポールの base_url です。北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions に置き換えてください
# バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/compatible-mode/v1/chat/completions に置き換えてください
# === 実行前にこのコメントを削除してください ===
curl -X POST https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions \
-H "Authorization: Bearer $DASHSCOPE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
    "model": "qwen-plus",
    "messages": [
        {
            "role": "user", 
            "content": "Who are you"
        }
    ],
    "stream": true,
    "stream_options": {
        "include_usage": true
    },
    "enable_thinking": true
}'

応答

data: {"choices":[{"delta":{"content":null,"role":"assistant","reasoning_content":""},"index":0,"logprobs":null,"finish_reason":null}],"object":"chat.completion.chunk","usage":null,"created":1745485391,"system_fingerprint":null,"model":"qwen-plus","id":"chatcmpl-e2edaf2c-8aaf-9e54-90e2-b21dd5045503"}

.....

data: {"choices":[{"finish_reason":"stop","delta":{"content":"","reasoning_content":null},"index":0,"logprobs":null}],"object":"chat.completion.chunk","usage":null,"created":1745485391,"system_fingerprint":null,"model":"qwen-plus","id":"chatcmpl-e2edaf2c-8aaf-9e54-90e2-b21dd5045503"}

data: {"choices":[],"object":"chat.completion.chunk","usage":{"prompt_tokens":10,"completion_tokens":360,"total_tokens":370},"created":1745485391,"system_fingerprint":null,"model":"qwen-plus","id":"chatcmpl-e2edaf2c-8aaf-9e54-90e2-b21dd5045503"}

data: [DONE]

サンプルコード

import os
from dashscope import Generation
import dashscope

# 以下はシンガポールリージョンの base_url です。バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/api/v1 に置き換えてください
# 北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/api/v1 に置き換えてください
dashscope.base_http_api_url = "https://dashscope-intl.aliyuncs.com/api/v1"

messages = [{"role": "user", "content": "Who are you?"}]

completion = Generation.call(
    # 環境変数を設定していない場合は、次の行を Model Studio API キーに置き換えます： api_key = "sk-xxx",
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model="qwen-plus",
    messages=messages,
    result_format="message",
    enable_thinking=True,
    stream=True,
    incremental_output=True,
)

# 完全な思考プロセスを定義。
reasoning_content = ""
# 完全な応答を定義。
answer_content = ""
# 思考プロセスが終了し、応答が開始されたかどうかを判定。
is_answering = False

print("=" * 20 + "Thinking process" + "=" * 20)

for chunk in completion:
    # 思考プロセスと応答の両方が空の場合、無視。
    if (
        chunk.output.choices[0].message.content == ""
        and chunk.output.choices[0].message.reasoning_content == ""
    ):
        pass
    else:
        # 現在が思考プロセスの場合。
        if (
            chunk.output.choices[0].message.reasoning_content != ""
            and chunk.output.choices[0].message.content == ""
        ):
            print(chunk.output.choices[0].message.reasoning_content, end="", flush=True)
            reasoning_content += chunk.output.choices[0].message.reasoning_content
        # 現在が応答の場合。
        elif chunk.output.choices[0].message.content != "":
            if not is_answering:
                print("\n" + "=" * 20 + "Full response" + "=" * 20)
                is_answering = True
            print(chunk.output.choices[0].message.content, end="", flush=True)
            answer_content += chunk.output.choices[0].message.content

# 完全な思考プロセスと応答を出力するには、次のコードのコメントを外して実行してください。
# print("=" * 20 + "Full thinking process" + "=" * 20 + "\n")
# print(f"{reasoning_content}")
# print("=" * 20 + "Full response" + "=" * 20 + "\n")
# print(f"{answer_content}")

応答

====================Thinking process====================
Okay, the user is asking, "Who are you?" I need to answer this question. First, I must clarify my identity: I am Qwen, a large-scale language model developed by Alibaba Cloud. Next, I need to explain my functions and uses, such as answering questions, creating text, and logical reasoning. I should also emphasize that my goal is to be a helpful assistant to the user, providing help and support.

When responding, I should maintain a conversational tone and avoid technical jargon or complex sentences. I can add friendly phrases, like "Hello there!~", to make the conversation more natural. Also, I must ensure the information is accurate and does not omit key points, such as my developer, main functions, and use cases.

I also need to consider potential follow-up questions from the user, such as specific application examples or technical details. So, I can subtly plant seeds in my response to encourage further questions. For example, mentioning "Whether it's a question about daily life or a professional topic, I can do my best to help" is both comprehensive and open-ended.

Finally, I will check if the response is fluent and free of repetitive or redundant information, ensuring it is concise and clear. I will also maintain a balance between being friendly and professional, so the user finds me both approachable and reliable.
====================Full response====================
Hello there!~ I am Qwen, a large-scale language model developed by Alibaba Cloud. I can answer questions, create text, perform logical reasoning, write code, and more, with the goal of providing help and support to users. Whether you have questions about daily life or professional topics, I will do my best to assist. How can I help you?

サンプルコード

// DashScope SDK バージョン >= 2.19.4
import java.util.Arrays;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.alibaba.dashscope.aigc.generation.Generation;
import com.alibaba.dashscope.aigc.generation.GenerationParam;
import com.alibaba.dashscope.aigc.generation.GenerationResult;
import com.alibaba.dashscope.common.Message;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.InputRequiredException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import io.reactivex.Flowable;
import java.lang.System;
import com.alibaba.dashscope.utils.Constants;

public class Main {
    static {
        // 以下はシンガポールリージョンの base_url です。バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/api/v1 に置き換えてください
        // 北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/api/v1 に置き換えてください
        Constants.baseHttpApiUrl="https://dashscope-intl.aliyuncs.com/api/v1";
    }
    private static final Logger logger = LoggerFactory.getLogger(Main.class);
    private static StringBuilder reasoningContent = new StringBuilder();
    private static StringBuilder finalContent = new StringBuilder();
    private static boolean isFirstPrint = true;

    private static void handleGenerationResult(GenerationResult message) {
        String reasoning = message.getOutput().getChoices().get(0).getMessage().getReasoningContent();
        String content = message.getOutput().getChoices().get(0).getMessage().getContent();

        if (!reasoning.isEmpty()) {
            reasoningContent.append(reasoning);
            if (isFirstPrint) {
                System.out.println("====================Thinking process====================");
                isFirstPrint = false;
            }
            System.out.print(reasoning);
        }

        if (!content.isEmpty()) {
            finalContent.append(content);
            if (!isFirstPrint) {
                System.out.println("\n====================Full response====================");
                isFirstPrint = true;
            }
            System.out.print(content);
        }
    }
    private static GenerationParam buildGenerationParam(Message userMsg) {
        return GenerationParam.builder()
                // 環境変数を設定していない場合は、次の行を Model Studio API キーに置き換えます： .apiKey("sk-xxx")
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-plus")
                .enableThinking(true)
                .incrementalOutput(true)
                .resultFormat("message")
                .messages(Arrays.asList(userMsg))
                .build();
    }
    public static void streamCallWithMessage(Generation gen, Message userMsg)
            throws NoApiKeyException, ApiException, InputRequiredException {
        GenerationParam param = buildGenerationParam(userMsg);
        Flowable<GenerationResult> result = gen.streamCall(param);
        result.blockingForEach(message -> handleGenerationResult(message));
    }

    public static void main(String[] args) {
        try {
            Generation gen = new Generation();
            Message userMsg = Message.builder().role(Role.USER.getValue()).content("Who are you?").build();
            streamCallWithMessage(gen, userMsg);
//             最終結果を出力。
//            if (reasoningContent.length() > 0) {
//                System.out.println("\n====================Full response====================");
//                System.out.println(finalContent.toString());
//            }
        } catch (ApiException | NoApiKeyException | InputRequiredException e) {
            logger.error("An exception occurred: {}", e.getMessage());
        }
        System.exit(0);
    }
}

戻り値

====================Thinking process====================
Okay, the user is asking "Who are you?", and I need to answer based on my predefined settings. First, my role is Qwen, a large-scale language model from Alibaba Group. I should keep the tone conversational, simple, and easy to understand.

The user might be new to me or wants to confirm my identity. I should first state who I am directly, then briefly explain my functions and uses, such as answering questions, creating text, and coding. I should also mention my multilingual support so the user knows I can handle requests in different languages.

Also, according to the guidelines, I should maintain a human-like persona, so the tone should be friendly. I might use emojis to add a touch of warmth. At the same time, I might need to guide the user to ask more questions or use my features, for example, by asking what they need help with.

I need to be careful not to use complex terminology and avoid being verbose. I will check for any missed key points, such as multilingual support and specific capabilities. I must ensure the response meets all requirements, including being conversational and concise.
====================Full response====================
Hello! I am Qwen, a large-scale language model from Alibaba Group. I can answer questions and create text, such as stories, official documents, emails, and playbooks. I can also perform logical reasoning, write code, express opinions, and play games. I am proficient in multiple languages, including but not limited to Chinese, English, German, French, and Spanish. Is there anything I can help you with?

サンプルコード

# ======= 重要 =======
# 以下はシンガポールリージョンの URL です。北京リージョンのモデルを使用する場合は、URL を https://dashscope.aliyuncs.com/api/v1/services/aigc/text-generation/generation に置き換えてください
# バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/api/v1/services/aigc/text-generation/generation に置き換えてください
# === 実行前にこのコメントを削除してください ===
curl -X POST "https://dashscope-intl.aliyuncs.com/api/v1/services/aigc/text-generation/generation" \
-H "Authorization: Bearer $DASHSCOPE_API_KEY" \
-H "Content-Type: application/json" \
-H "X-DashScope-SSE: enable" \
-d '{
    "model": "qwen-plus",
    "input":{
        "messages":[      
            {
                "role": "user",
                "content": "Who are you?"
            }
        ]
    },
    "parameters":{
        "enable_thinking": true,
        "incremental_output": true,
        "result_format": "message"
    }
}'

応答

id:1
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"","reasoning_content":"Hmm","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":14,"input_tokens":11,"output_tokens":3},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:2
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"","reasoning_content":",","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":15,"input_tokens":11,"output_tokens":4},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:3
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"","reasoning_content":"the user","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":16,"input_tokens":11,"output_tokens":5},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:4
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"","reasoning_content":" asks","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":17,"input_tokens":11,"output_tokens":6},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:5
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"","reasoning_content":" '","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":18,"input_tokens":11,"output_tokens":7},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}
......

id:358
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"help","reasoning_content":"","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":373,"input_tokens":11,"output_tokens":362},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:359
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":",","reasoning_content":"","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":374,"input_tokens":11,"output_tokens":363},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:360
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":" feel free","reasoning_content":"","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":375,"input_tokens":11,"output_tokens":364},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:361
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":" to","reasoning_content":"","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":376,"input_tokens":11,"output_tokens":365},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:362
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":" let me know","reasoning_content":"","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":377,"input_tokens":11,"output_tokens":366},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:363
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"!","reasoning_content":"","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":378,"input_tokens":11,"output_tokens":367},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

id:364
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"","reasoning_content":"","role":"assistant"},"finish_reason":"stop"}]},"usage":{"total_tokens":378,"input_tokens":11,"output_tokens":367},"request_id":"25d58c29-c47b-9e8d-a0f1-d6c309ec58b1"}

サンプルコード

from openai import OpenAI
import os

# OpenAI クライアントを初期化
client = OpenAI(
    # 環境変数を設定していない場合は、次の内容を Model Studio API キーに置き換えます： api_key="sk-xxx"
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    # 以下はシンガポールリージョンの base_url です。バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/compatible-mode/v1 に置き換えてください # 北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/compatible-mode/v1 に置き換えてください
    # 北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/compatible-mode/v1 に置き換えてください
    base_url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
)

messages = [{"role": "user", "content": "Who are you"}]

completion = client.chat.completions.create(
    model="qwen-plus",
    messages=messages,
    # enable_thinking パラメーターは思考プロセスを有効化し、thinking_budget パラメーターは推論プロセスの最大トークン数を設定します。
    extra_body={
        "enable_thinking": True,
        "thinking_budget": 50
        },
    stream=True,
    stream_options={
        "include_usage": True
    },
)

reasoning_content = ""  # 完全な思考プロセス
answer_content = ""  # 完全な応答
is_answering = False  # 応答フェーズが開始されたかどうかを示す
print("\n" + "=" * 20 + "Thinking process" + "=" * 20 + "\n")

for chunk in completion:
    if not chunk.choices:
        print("\nUsage:")
        print(chunk.usage)
        continue

    delta = chunk.choices[0].delta

    # 思考コンテンツのみを収集
    if hasattr(delta, "reasoning_content") and delta.reasoning_content is not None:
        if not is_answering:
            print(delta.reasoning_content, end="", flush=True)
        reasoning_content += delta.reasoning_content

    # コンテンツを受信した場合、応答を開始
    if hasattr(delta, "content") and delta.content:
        if not is_answering:
            print("\n" + "=" * 20 + "Full response" + "=" * 20 + "\n")
            is_answering = True
        print(delta.content, end="", flush=True)
        answer_content += delta.content

応答

====================Thinking process====================

Okay, the user asked "Who are you", so I need to give a clear and friendly response. First, I should state my identity, which is Qwen, developed by the Tongyi Lab under Alibaba Group. Next, I should explain my main functions, such as answering
====================Full response====================

I am Qwen, a large-scale language model developed by the Tongyi Lab at Alibaba Group. I can answer questions, create text, perform logical reasoning, and write code, with the goal of providing help and convenience to users. How can I help you?

サンプルコード

import OpenAI from "openai";
import process from 'process';

// OpenAI クライアントを初期化します
const openai = new OpenAI({
    apiKey: process.env.DASHSCOPE_API_KEY, // 環境変数から読み取ります
    // 以下は、シンガポールリージョンの base_url です。バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/compatible-mode/v1 に置き換えます。
    // 北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/compatible-mode/v1 に置き換えます。
    baseURL: 'https://dashscope-intl.aliyuncs.com/compatible-mode/v1'
});

let reasoningContent = '';
let answerContent = '';
let isAnswering = false;


async function main() {
    try {
        const messages = [{ role: 'user', content: 'Who are you' }];
        const stream = await openai.chat.completions.create({
            model: 'qwen-plus',
            messages,
            stream: true,
            // enable_thinking パラメーターは推論プロセスを有効化し、thinking_budget パラメーターは推論プロセスの最大トークン数を設定します。
            enable_thinking: true,
            thinking_budget: 50
        });
        console.log('\n' + '='.repeat(20) + '推論プロセス' + '='.repeat(20) + '\n');

        for await (const chunk of stream) {
            if (!chunk.choices?.length) {
                console.log('\nUsage:');
                console.log(chunk.usage);
                continue;
            }

            const delta = chunk.choices[0].delta;
            
            // 推論コンテンツのみを収集します
            if (delta.reasoning_content !== undefined && delta.reasoning_content !== null) {
                if (!isAnswering) {
                    process.stdout.write(delta.reasoning_content);
                }
                reasoningContent += delta.reasoning_content;
            }

            // コンテンツを受信すると、応答を開始します
            if (delta.content !== undefined && delta.content) {
                if (!isAnswering) {
                    console.log('\n' + '='.repeat(20) + '完全な応答' + '='.repeat(20) + '\n');
                    isAnswering = true;
                }
                process.stdout.write(delta.content);
                answerContent += delta.content;
            }
        }
    } catch (error) {
        console.error('Error:', error);
    }
}

main();

応答

====================Thinking process====================

Okay, the user asked "Who are you", so I need to give a clear and accurate response. First, I should introduce my identity, which is Qwen, developed by the Tongyi Lab under Alibaba Group. Next, I should explain my main functions, such as answering questions
====================Full response====================

I am Qwen, an ultra-large language model independently developed by the Tongyi Lab under Alibaba Group. I can perform various tasks such as answering questions, creating text, logical reasoning, and coding. If you have any questions or need help, feel free to let me know!

サンプルコード

# ======= 重要 =======
# 以下はシンガポールの base_url です。北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions に置き換えてください
# バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/compatible-mode/v1/chat/completions に置き換えてください
# === 実行前にこのコメントを削除してください ===
curl -X POST https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions \
-H "Authorization: Bearer $DASHSCOPE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
    "model": "qwen-plus",
    "messages": [
        {
            "role": "user", 
            "content": "Who are you"
        }
    ],
    "stream": true,
    "stream_options": {
        "include_usage": true
    },
    "enable_thinking": true,
    "thinking_budget": 50
}'

応答

data: {"choices":[{"delta":{"content":null,"role":"assistant","reasoning_content":""},"index":0,"logprobs":null,"finish_reason":null}],"object":"chat.completion.chunk","usage":null,"created":1745485391,"system_fingerprint":null,"model":"qwen-plus","id":"chatcmpl-e2edaf2c-8aaf-9e54-90e2-b21dd5045503"}

.....

data: {"choices":[{"finish_reason":"stop","delta":{"content":"","reasoning_content":null},"index":0,"logprobs":null}],"object":"chat.completion.chunk","usage":null,"created":1745485391,"system_fingerprint":null,"model":"qwen-plus","id":"chatcmpl-e2edaf2c-8aaf-9e54-90e2-b21dd5045503"}

data: {"choices":[],"object":"chat.completion.chunk","usage":{"prompt_tokens":10,"completion_tokens":360,"total_tokens":370},"created":1745485391,"system_fingerprint":null,"model":"qwen-plus","id":"chatcmpl-e2edaf2c-8aaf-9e54-90e2-b21dd5045503"}

data: [DONE]

サンプルコード

import os
from dashscope import Generation
import dashscope
# 以下はシンガポールリージョンの base_url です。バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/api/v1 に置き換えてください
# 北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/api/v1 に置き換えてください
dashscope.base_http_api_url = "https://dashscope-intl.aliyuncs.com/api/v1/"

messages = [{"role": "user", "content": "Who are you?"}]

completion = Generation.call(
    # 環境変数を設定していない場合は、次の行を Model Studio API キーに置き換えます： api_key = "sk-xxx",
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model="qwen-plus",
    messages=messages,
    result_format="message",
    enable_thinking=True,
    # 推論プロセスの最大トークン数を設定。
    thinking_budget=50,
    stream=True,
    incremental_output=True,
)

# 完全な思考プロセスを定義。
reasoning_content = ""
# 完全な応答を定義。
answer_content = ""
# 思考プロセスが終了し、応答が開始されたかどうかを判定。
is_answering = False

print("=" * 20 + "Thinking process" + "=" * 20)

for chunk in completion:
    # 思考プロセスと応答の両方が空の場合、無視。
    if (
        chunk.output.choices[0].message.content == ""
        and chunk.output.choices[0].message.reasoning_content == ""
    ):
        pass
    else:
        # 現在が思考プロセスの場合。
        if (
            chunk.output.choices[0].message.reasoning_content != ""
            and chunk.output.choices[0].message.content == ""
        ):
            print(chunk.output.choices[0].message.reasoning_content, end="", flush=True)
            reasoning_content += chunk.output.choices[0].message.reasoning_content
        # 現在が応答フェーズの場合。
        elif chunk.output.choices[0].message.content != "":
            if not is_answering:
                print("\n" + "=" * 20 + "Full response" + "=" * 20)
                is_answering = True
            print(chunk.output.choices[0].message.content, end="", flush=True)
            answer_content += chunk.output.choices[0].message.content

# 完全な思考プロセスと完全な応答を出力するには、次のコードのコメントを外して実行してください。
# print("=" * 20 + "Full thinking process" + "=" * 20 + "\n")
# print(f"{reasoning_content}")
# print("=" * 20 + "Full response" + "=" * 20 + "\n")
# print(f"{answer_content}")

戻り値

====================Thinking process====================
Okay, the user is asking "Who are you?", so I need to give a clear and friendly response. First, I should introduce my identity, which is Qwen, developed by the Tongyi Lab under Alibaba Group. Next, I should explain my main functions, such as
====================Full response====================
I am Qwen, a large-scale language model independently developed by the Tongyi Lab at Alibaba Group. I can answer questions, create text, perform logical reasoning, and write code, with the goal of providing users with comprehensive, accurate, and useful information and assistance. How can I help you?

サンプルコード

// DashScope SDK バージョン >= 2.19.4
import java.util.Arrays;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.alibaba.dashscope.aigc.generation.Generation;
import com.alibaba.dashscope.aigc.generation.GenerationParam;
import com.alibaba.dashscope.aigc.generation.GenerationResult;
import com.alibaba.dashscope.common.Message;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.InputRequiredException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import io.reactivex.Flowable;
import java.lang.System;
import com.alibaba.dashscope.utils.Constants;

public class Main {
    static {
        // 以下はシンガポールリージョンの base_url です。バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/api/v1 に置き換えてください
        // 北京リージョンのモデルを使用する場合は、base_url を https://dashscope.aliyuncs.com/api/v1 に置き換えてください
        Constants.baseHttpApiUrl="https://dashscope-intl.aliyuncs.com/api/v1";
    }
    private static final Logger logger = LoggerFactory.getLogger(Main.class);
    private static StringBuilder reasoningContent = new StringBuilder();
    private static StringBuilder finalContent = new StringBuilder();
    private static boolean isFirstPrint = true;

    private static void handleGenerationResult(GenerationResult message) {
        String reasoning = message.getOutput().getChoices().get(0).getMessage().getReasoningContent();
        String content = message.getOutput().getChoices().get(0).getMessage().getContent();

        if (!reasoning.isEmpty()) {
            reasoningContent.append(reasoning);
            if (isFirstPrint) {
                System.out.println("====================Thinking process====================");
                isFirstPrint = false;
            }
            System.out.print(reasoning);
        }

        if (!content.isEmpty()) {
            finalContent.append(content);
            if (!isFirstPrint) {
                System.out.println("\n====================Full response====================");
                isFirstPrint = true;
            }
            System.out.print(content);
        }
    }
    private static GenerationParam buildGenerationParam(Message userMsg) {
        return GenerationParam.builder()
                // 環境変数を設定していない場合は、次の行を Model Studio API キーに置き換えます： .apiKey("sk-xxx")
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-plus")
                .enableThinking(true)
                .thinkingBudget(50)
                .incrementalOutput(true)
                .resultFormat("message")
                .messages(Arrays.asList(userMsg))
                .build();
    }
    public static void streamCallWithMessage(Generation gen, Message userMsg)
            throws NoApiKeyException, ApiException, InputRequiredException {
        GenerationParam param = buildGenerationParam(userMsg);
        Flowable<GenerationResult> result = gen.streamCall(param);
        result.blockingForEach(message -> handleGenerationResult(message));
    }

    public static void main(String[] args) {
        try {
            Generation gen = new Generation();
            Message userMsg = Message.builder().role(Role.USER.getValue()).content("Who are you?").build();
            streamCallWithMessage(gen, userMsg);
//             最終結果を出力。
//            if (reasoningContent.length() > 0) {
//                System.out.println("\n====================Full response====================");
//                System.out.println(finalContent.toString());
//            }
        } catch (ApiException | NoApiKeyException | InputRequiredException e) {
            logger.error("An exception occurred: {}", e.getMessage());
        }
        System.exit(0);
    }
}

====================Thinking process====================
Okay, the user is asking "Who are you?", so I need to give a clear and friendly response. First, I should introduce my identity, which is Qwen, developed by the Tongyi Lab under Alibaba Group. Next, I should explain my main functions, such as
====================Full response====================
I am Qwen, a large-scale language model independently developed by the Tongyi Lab at Alibaba Group. I can answer questions, create text, perform logical reasoning, and write code, with the goal of providing users with comprehensive, accurate, and useful information and assistance. How can I help you?

サンプルコード

# ======= 重要 =======
# 以下はシンガポールリージョンの URL です。北京リージョンのモデルを使用する場合は、URL を https://dashscope.aliyuncs.com/api/v1/services/aigc/text-generation/generation に置き換えてください
# バージニアリージョンのモデルを使用する場合は、base_url を https://dashscope-us.aliyuncs.com/api/v1/services/aigc/text-generation/generation に置き換えてください
# === 実行前にこのコメントを削除してください ===
curl -X POST "https://dashscope-intl.aliyuncs.com/api/v1/services/aigc/text-generation/generation" \
-H "Authorization: Bearer $DASHSCOPE_API_KEY" \
-H "Content-Type: application/json" \
-H "X-DashScope-SSE: enable" \
-d '{
    "model": "qwen-plus",
    "input":{
        "messages":[      
            {
                "role": "user",
                "content": "Who are you?"
            }
        ]
    },
    "parameters":{
        "enable_thinking": true,
        "thinking_budget": 50,
        "incremental_output": true,
        "result_format": "message"
    }
}'

戻り値

id:1
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"","reasoning_content":"Okay","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":14,"output_tokens":3,"input_tokens":11,"output_tokens_details":{"reasoning_tokens":1}},"request_id":"2ce91085-3602-9c32-9c8b-fe3d583a2c38"}

id:2
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"","reasoning_content":",","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":15,"output_tokens":4,"input_tokens":11,"output_tokens_details":{"reasoning_tokens":2}},"request_id":"2ce91085-3602-9c32-9c8b-fe3d583a2c38"}

......

id:133
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"!","reasoning_content":"","role":"assistant"},"finish_reason":"null"}]},"usage":{"total_tokens":149,"output_tokens":138,"input_tokens":11,"output_tokens_details":{"reasoning_tokens":50}},"request_id":"2ce91085-3602-9c32-9c8b-fe3d583a2c38"}

id:134
event:result
:HTTP_STATUS/200
data:{"output":{"choices":[{"message":{"content":"","reasoning_content":"","role":"assistant"},"finish_reason":"stop"}]},"usage":{"total_tokens":149,"output_tokens":138,"input_tokens":11,"output_tokens_details":{"reasoning_tokens":50}},"request_id":"2ce91085-3602-9c32-9c8b-fe3d583a2c38"}