Alibaba Cloud Model Studio:DeepSeek

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(
    # 環境変数が設定されていない場合、以下を Model Studio API キーに置き換えます: api_key="sk-xxx"
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
)

messages = [{"role": "user", "content": "Who are you"}]
completion = client.chat.completions.create(
    model="deepseek-v3.2",
    messages=messages,
    # extra_body に enable_thinking を設定して思考モードを有効化
    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フラグ
print("\n" + "=" * 20 + "Thinking process" + "=" * 20 + "\n")

for chunk in completion:
    if not chunk.choices:
        print("\n" + "=" * 20 + "Token usage" + "=" * 20 + "\n")
        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時点で応答を開始
    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====================

Ah, the user is asking who I am. This is a very common opening question. I need to introduce my identity and functions simply and clearly. I can start with my company background and core capabilities to help the user quickly understand.
I should highlight my free-to-use nature and text-based strengths, but avoid going into too much detail. Finally, I'll guide the conversation with an open-ended question, which is in line with the nature of an assistant.
I'll position myself as an enterprise-level AI assistant, which is both professional and friendly. The emoji in parentheses can add a touch of friendliness.
====================Full response====================

Hello! I am DeepSeek, an AI assistant created by DeepSeek.

I am a text-only model. Although I do not support multimodal recognition, I have a file upload feature that can help you process various files such as images, txt, pdf, ppt, word, and excel, and read text information from them to assist you. I am completely free to use, have a 128K context window, and support web search (you need to manually enable it in the Web/App).

My knowledge is current up to July 2024, and I will help you with enthusiasm and care. You can download my app from the official app store.

Is there anything I can help you with? Whether it's a question about your studies, work, or daily life, I'm happy to assist you! ✨
====================Token usage====================

CompletionUsage(completion_tokens=238, prompt_tokens=5, total_tokens=243, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=93, rejected_prediction_tokens=None), prompt_tokens_details=None)

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({
    // 環境変数が設定されていない場合、以下を Model Studio API キーに置き換えます: apiKey: "sk-xxx"
    apiKey: process.env.DASHSCOPE_API_KEY, 
    baseURL: 'https://dashscope.aliyuncs.com/compatible-mode/v1'
});

let reasoningContent = ''; // 完全な思考プロセス
let answerContent = ''; // 完全な応答
let isAnswering = false; // 応答フェーズが開始されたかどうかを示すフラグ

async function main() {
    try {
        const messages = [{ role: 'user', content: 'Who are you' }];
        
        const stream = await openai.chat.completions.create({
            model: 'deepseek-v3.2',
            messages,
            // 注: Node.js SDK では、enable_thinking のような非標準パラメーターはトップレベルのプロパティとして渡され、extra_body に配置する必要はありません。
            enable_thinking: true,
            stream: true,
            stream_options: {
                include_usage: true
            },
        });

        console.log('\n' + '='.repeat(20) + 'Thinking process' + '='.repeat(20) + '\n');

        for await (const chunk of stream) {
            if (!chunk.choices?.length) {
                console.log('\n' + '='.repeat(20) + 'Token usage' + '='.repeat(20) + '\n');
                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時点で応答を開始
            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();

応答

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

Ah, the user is asking who I am. This is a very common opening question. I need to introduce my identity and core functions simply and clearly, without going into too much detail.

I can start with my company background and basic positioning, then list a few key capabilities to let the user quickly understand what I can do. I'll end with an open-ended question to make it easy for the user to continue.

I should highlight practical features like being free, having a long context, and file processing. I'll maintain a friendly but restrained tone, without using emojis.
====================Full response====================

Hello! I am DeepSeek, an AI assistant created by DeepSeek.

I am a text-only model with a 128K context window, and I can help you answer questions, engage in conversations, and assist with text-based tasks. Although I do not support multimodal recognition, I can process files you upload, such as images, txt, pdf, ppt, word, and excel, and read text information from them to help you.

I am completely free to use and have no voice function, but you can download my app from the official app store. To use web search, remember to manually enable it in the Web or App.

My knowledge is current up to July 2024, and I will help you with enthusiasm and care. If you have any questions or need assistance, just let me know! I'm happy to help. ✨
====================Token usage====================

{
  prompt_tokens: 5,
  completion_tokens: 243,
  total_tokens: 248,
  completion_tokens_details: { reasoning_tokens: 83 }
}

サンプルコード

curl -X POST https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions \
-H "Authorization: Bearer $DASHSCOPE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
    "model": "deepseek-v3.2",
    "messages": [
        {
            "role": "user", 
            "content": "Who are you"
        }
    ],
    "stream": true,
    "stream_options": {
        "include_usage": true
    },
    "enable_thinking": true
}'

サンプルコード

import os
from dashscope import Generation

# リクエストパラメーターを初期化
messages = [{"role": "user", "content": "Who are you?"}]

completion = Generation.call(
    # 環境変数が設定されていない場合、以下を Model Studio API キーに置き換えます: api_key="sk-xxx"
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model="deepseek-v3.2",
    messages=messages,
    result_format="message",  # 結果フォーマットを message に設定
    enable_thinking=True,
    stream=True,              # ストリーミング出力を有効化
    incremental_output=True,  # 増分出力を有効化
)

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フラグ

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

for chunk in completion:
    message = chunk.output.choices[0].message
    # 思考コンテンツのみを収集
    if "reasoning_content" in message:
        if not is_answering:
            print(message.reasoning_content, end="", flush=True)
        reasoning_content += message.reasoning_content

    # コンテンツを受信した時点で応答を開始
    if message.content:
        if not is_answering:
            print("\n" + "=" * 20 + "Full response" + "=" * 20 + "\n")
            is_answering = True
        print(message.content, end="", flush=True)
        answer_content += message.content

print("\n" + "=" * 20 + "Token usage" + "=" * 20 + "\n")
print(chunk.usage)

応答

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

Oh, the user is asking who I am. This is a very basic self-introduction question. I need to state my identity and functions concisely and clearly, avoiding complexity. I can start with my company background and core capabilities to help the user quickly understand.
Considering the user might be new, I can add some typical use cases and features, such as being free, having a long context, and file processing. I'll end with an open-ended invitation for help, maintaining a friendly attitude.
No need for too many technical details, the focus should be on ease of use and practicality.
====================Full response====================

Hello! I am DeepSeek, an AI assistant created by DeepSeek.

I am a text-only model. Although I do not support multimodal recognition, I have a file upload feature that can help you process files like images, txt, pdf, ppt, word, and excel by reading the text information for analysis. I am completely free to use, have a 128K context window, and support web search (you need to manually enable it).

My knowledge is current up to July 2024, and I will help you with enthusiasm and care. You can download my app from the official app store.

If you have any questions or need help, just ask! I'm happy to answer your questions and assist with various tasks. ✨
====================Token usage====================

{"input_tokens": 6, "output_tokens": 240, "total_tokens": 246, "output_tokens_details": {"reasoning_tokens": 92}}

サンプルコード

// DashScope SDK のバージョンは 2.19.4 以降である必要があります。
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 java.util.Arrays;

public class Main {
    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 != null && !reasoning.isEmpty()) {
            reasoningContent.append(reasoning);
            if (isFirstPrint) {
                System.out.println("====================Thinking process====================");
                isFirstPrint = false;
            }
            System.out.print(reasoning);
        }
        if (content != null && !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()
                // 環境変数が設定されていない場合、以下の行を .apiKey("sk-xxx") に置き換えます
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("deepseek-v3.2")
                .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);
        } catch (ApiException | NoApiKeyException | InputRequiredException e) {
            System.err.println("An exception occurred: " + e.getMessage());
        }
    }
}

応答

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

Hmm, the user is asking a simple self-introduction question. This is a common query, so I need to state my identity and function clearly and quickly. I'll use a relaxed and friendly tone to introduce myself as DeepSeek-V3, created by DeepSeek. I can also mention the types of help I can provide, such as answering questions, chatting, and tutoring. Finally, I'll add an emoji to be more approachable. I should keep it concise and clear.
====================Full response====================

I am DeepSeek-V3, an intelligent assistant created by DeepSeek! I can help you answer various questions, provide suggestions, look up information, and even chat with you! Feel free to ask me anything about your studies, work, or daily life. How can I help you?

サンプルコード

curl -X POST "https://dashscope.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": "deepseek-v3.2",
    "input":{
        "messages":[      
            {
                "role": "user",
                "content": "Who are you?"
            }
        ]
    },
    "parameters":{
        "enable_thinking": true,
        "incremental_output": true,
        "result_format": "message"
    }
}'