What is Alibaba Cloud Model Studio - Alibaba Cloud Model Studio

Alibaba Cloud Model Studio integrates the complete Qwen model family and leading third-party models, covering multiple modalities and scenarios. Call models on demand without managing the underlying infrastructure, reducing your operational overhead.

api Call APIs

Chat with an LLM to generate content, summaries, and more using just a few lines of code.

Model Studio is compatible with OpenAI API specifications. Simply update the API key, base URL, and model name to migrate your existing OpenAI code to Model Studio.

import os
from openai import OpenAI

# Note: The base_url varies by region. The following example uses the base_url for the Singapore region.
# - Singapore: https://dashscope-intl.aliyuncs.com/compatible-mode/v1
# - China (Beijing): https://dashscope.aliyuncs.com/compatible-mode/v1
client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"), 
    base_url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1"
)
completion = client.chat.completions.create(
    model="qwen3.5-plus",
    messages=[{"role": "user", "content": "Who are you?"}]
)
print(completion.choices[0].message.content)

Model service

Model Studio provides out-of-the-box model services. Directly call Qwen models and third-party LLMs, such as DeepSeek and Kimi, with no deployment or maintenance required. Complete model list

Qwen flagship models:
- Qwen-Max: The best-performing model in the Qwen3 series, suitable for complex, multi-step tasks.
- Qwen-Plus: Offers a balance of performance, speed, and cost, making it the recommended choice for most scenarios.
  The latest Qwen3.5-Plus series excels at various tasks, including language understanding, logical reasoning, code generation, agent tasks, and image and video understanding. We highly recommend this series for its versatility.
- Qwen-Flash: Cost-effective and low-latency, suitable for simple tasks requiring fast responses.
- Qwen-Coder: Excels at tool calling and environment interaction, and is specialized for code generation and understanding.
Multimodal coverage: Includes capabilities such as text generation, visual understanding, image generation, video generation, speech recognition and synthesis, and embedding.
Domain-specific models: For specific industries and tasks, provides domain models for long-text processing, translation, data mining, intent recognition, role-playing, and deep research.

Billing

Activating Model Studio does not incur any fees. You incur costs only when you call models. See Billable items and the Model pricing.

Free quota for new users

Model Studio provides new users with an exclusive free quota in the Singapore region. After the quota is exhausted, billing switches to pay-as-you-go. To avoid unexpected charges, turn on the Free quota only feature. The service will automatically stop when the quota runs out.

Payment methods

Model calls are automatically charged on an hourly basis. For supported payment methods, see Introduction to payment methods.

View bills and usage

Billing details: Go to the Billing Details and Cost Analysis pages.
Call statistics: Approximately one hour after making a model call, go to the Monitoring (Singapore),Monitoring (Virginia), or Monitoring (Beijing) page, set the filter criteria, and click Monitor in the Actions column. Then, view statistics such as call volume, token consumption, and success rate for the model. See Monitoring.
Coding Plan usage: If you have subscribed to a Coding Plan, view the quota consumption on the Coding Plan page. Coding Plan uses a fixed monthly fee and provides a monthly request quota for use in AI coding tools. See Coding Plan overview.

Get started with Model Studio

Try models online: Playground (Singapore), Playground (Virginia), or Playground (Beijing)
Make your first API request: Make the first call to a Qwen API

FAQ

Q: Is my data secure? Will Model Studio use my data for training?

A: Alibaba Cloud is committed to protecting your data privacy and will never use your data for model training. Additionally, the data you transmit while building applications or training models is encrypted to ensure data security. See Privacy notice.

Q: What are the differences between the International (Singapore), US (Virginia), and Chinese mainland (Beijing) regions?

A: Model Studio provides model services in the following regions: International (Singapore), US (Virginia) and Chinese mainland (Beijing). Selecting a geographically nearby region for API calls reduces network latency. Endpoints (endpoint/base URL) vary by region, and API keys are not interchangeable. Supported models, platform features, and pricing also differ by region, see Models.

Q: How can I avoid automatic charges?

A: Model Studio uses a pay-as-you-go billing method and does not offer an option to avoid automatic charges. To avoid surprise bills:

Delete API keys: Go to the API-KEY (Singapore), API-KEY (Virginia), or API-KEY (Beijing) page and delete all API keys.
Stop all calls: Stop model calls in applications, agents, and workflows. Also, check for any scheduled tasks or background processes.
Turn on "Free quota only" (for new users within the validity period only): On the model details page, turn on this switch. The service will stop automatically when the free quota is used up, preventing it from switching to the paid mode. This applies only to international (Singapore) models and is valid only within the validity period.
Set up cost monitoring and alerts: View your Billing Details and Monitoring data, and set up high-spending alerts to receive timely notifications about unusual spending.
Subscribe to Coding Plan: This plan uses a fixed monthly fee and provides a monthly quota, eliminating the risk of pay-as-you-go charges. Make sure to use its exclusive base URL and API key. Otherwise, model calls will still incur charges. See Coding Plan overview.

Q: How to use Qwen3 or DeepSeek?

Try it online: Visit the Models (Singapore), Models (Virginia), or Models (Beijing) page. Click a model to try it out (DeepSeek is available only in the Beijing region).
Use API: See Make the first call to a Qwen API. For supported models, see Model list.
Use developer tools (such as Claude Code): See Connect to clients or developer tools.