Viewing a Built-in Model in Model Square

MaaS provides various open-source models. You can check them on the Model Square. The model details page shows all necessary information. You can choose suitable models for training and inference to incorporate into your enterprise systems.

Prerequisites

You have registered a HUAWEI ID and enabled Huawei Cloud services.

Billing

Calling deployed model services: Billing is based on compute. For details, see Compute Resource Billing Items.

Accessing the Model Square

Log in to the MaaS console and select the target region on the top navigation bar.
In the navigation pane, choose Model Square.

In the Filter area on the Model Square page, filter models by type, context length, advanced features, series, and supported jobs, or search by model name.

For details about model series, see Model Introduction. The supported filter criteria may vary depending on the region. The following uses CN-Hong Kong as an example.

**Table 1** Model filters
Filter	Description
Type	You can filter models by type, including text generation and image understanding. If you select multiple model types, the page displays the collection of the selected model types.
Context Length	You can filter models by context length, including 128K, 64K, 32K, 16K, and no more than 8K. If you select multiple context lengths, the page displays the model set of the selected context lengths.
Advanced Capabilities	You can filter models based on their ability to handle function calls or deep reasoning.
Models	You can filter models by series, such as DeepSeek, Qwen, GLM, and DeepSeek Coder. Models supported by each region are different. For details, see Model Introduction. If you select multiple models, the page displays the collection of the selected model series.
Supported Job Types	You can filter models by their job types, such as deployment.

Perform the following operations on the target model card on the Model Square page:

Model Introduction

The following table lists the models supported by the MaaS platform. For details about the models, go to the model details page.

**Table 2** Models in the Model Square
Model Series		Type	Use Case	Supported Language	Supported Region	Model Introduction
GLM	GLM-5.2	Text generation	Reasoning, coding, and agentic tasks	Chinese and English	CN-Hong Kong	GLM-5.2 is the next-generation flagship model from Zhipu AI, open-sourced on June 17, 2026. With state-of-the-art open-source coding and long-horizon task capabilities, it significantly enhances engineering stability in real-world development and improves multi-platform delivery reliability.
	GLM-5.1	Text generation	Reasoning, coding, and agentic tasks	Chinese and English	CN-Hong Kong	GLM-5.1 is the latest flagship model from Zhipu. It provides enhanced coding capabilities and improved performance in long-horizon tasks. It can autonomously work for up to 8 hours in a single task, covering the entire process from planning and execution to iterative optimization and delivering engineering-level results. While GLM-5.1 matches Claude Opus 4.6 in general intelligence and raw coding proficiency, it significantly outperforms the global frontier in long-horizon sustained execution. It excels at autonomous, multi-stage tasks over extended periods, making it the ideal foundation for building highly resilient autonomous agents and long-horizon coding engines.
	GLM-5	Text generation	Reasoning, coding, and agentic tasks	Chinese and English	CN-Hong Kong	Compared with GLM-4.5, GLM-5 expands total parameters from 355B to 744B (boosting active parameters from 32B to 40B) and increases pre-training data from 23T to 28.5T tokens. In addition, GLM-5 integrates the DeepSeek sparse attention (DSA) mechanism, which significantly reduces deployment costs while maintaining the long context capability. Compared to GLM-4.7, GLM-5 delivers substantial performance gains across a wide spectrum of academic benchmarks. It establishes a new state-of-the-art (SOTA) among global open-source models in reasoning, coding, and agentic tasks, further closing the capability gap with frontier closed-source models.
DeepSeek	DeepSeek-V4-Pro	Text generation	Q&A and text generation inference	Chinese and English	CN-Hong Kong	DeepSeek-V4-Pro is the flagship version of the DeepSeek-V4 series. It adopts a Mixture-of-Experts (MoE) architecture, with trillions of parameters and an ultra-long 1M (one million token) context window.
	DeepSeek-V4-Flash	Text generation	Q&A and text generation inference	Chinese and English	CN-Hong Kong	DeepSeek-V4-Flash is a lightweight, efficient version of the DeepSeek-V4 series. It supports a 1M-token context window while offering faster, cheaper API services through a smaller model size and fewer active parameters.
	DeepSeek-R1-0528	Text generation	Q&A and text generation inference	Chinese and English	CN-Hong Kong	A DeepSeek-R1 agent model. It uses advanced technology for long-context understanding and fast inference. The model supports multi-modal interactions and API integrations. It enhances applications like intelligent customer service and data analytics, offering top cost-effectiveness for intelligent upgrades of enterprises.
	DeepSeek-V3	Text generation	Q&A and translation	Chinese and English	CN-Hong Kong	DeepSeek-V3 is a strong MoE language model. It uses a new load balancing method without extra loss and aims for better performance with multi-token predictions.
	DeepSeek-V3.1	Text generation	Q&A	Chinese and English	CN-Hong Kong	DeepSeek-V3.1 is a hybrid model with thinking and non-thinking modes. It matches DeepSeek-R1-0528's performance but offers quicker responses and better tool optimization.
	DeepSeek-V3.2	Text generation	Q&A	Chinese and English	CN-Hong Kong	DeepSeek-V3.2 achieves an exceptional balance between computational efficiency and advanced reasoning and agent capabilities, with overall performance reaching the level of GPT-5.
	DeepSeek-R1-Distill-Qwen-14B	Text generation	Q&A and text generation inference	Chinese and English	CN-Hong Kong	Qwen-14B is distilled from DeepSeek-R1 outputs and delivers capabilities on par with OpenAI's o1-mini. DeepSeek-R1 performs similarly to OpenAI-o1 in math, coding, and inference tasks.
	DeepSeek-R1-Distill-Qwen-32B	Text generation	Q&A and text generation inference	Chinese and English	CN-Hong Kong	Qwen-32B is distilled from DeepSeek-R1 outputs and delivers capabilities on par with OpenAI's o1-mini. DeepSeek-R1 performs similarly to OpenAI-o1 in math, coding, and inference tasks.
Deepseek-Coder		Text generation	Q&A and text generation inference	Chinese and English	CN-Hong Kong	DeepSeek Coder includes several code language models. Every model trains from scratch using 2 trillion tokens, with 87% being code and 13% English or Chinese text. It excels in multiple programming languages and performs well across various benchmarks among open-source code models.
Qwen	QwQ	Text generation	Q&A	English	CN-Hong Kong	QwQ is part of the Tongyi series of inference models. Unlike standard instruction-tuning models, QwQ excels at thinking and reasoning, delivering better results on complex tasks.
Qwen 2.5	Qwen 2.5	Text generation	Multilingual processing, mathematical inference, and Q&A	Chinese and English	CN-Hong Kong	Alibaba Cloud's Qwen 2.5 is a new addition to the Qwen series of LLMs. Qwen 2.5 offers various base and instruction-tuned language models, spanning sizes from 0.5 billion to 72 billion parameters.
Qwen 2.5	Qwen2.5-VL	Image understanding	Image understanding and Q&A	Chinese and English	CN-Hong Kong	Qwen-2.5-VL is an open-source multimodal visual language model created by Alibaba Cloud's Qwen team. It excels in visual and language understanding.
Qwen3	Qwen3	Text generation	Q&A	Chinese and English	CN-Hong Kong	The Qwen3 series includes LLMs and multimodal models created by the Qwen team. These models undergo extensive training using vast amounts of language and multimodal data, followed by fine-tuning with top-tier datasets.

Viewing Built-in Model Details

Log in to the MaaS console and select the target region on the top navigation bar.
In the navigation pane, choose Model Square.
In the Filter area on the Model Square page, filter models by type, context length, advanced features, series, and supported jobs, or search by model name.
Click a model card. On the Model Details page, view the model details, such as the model introduction, version information, and supported capabilities. In the upper right corner of the Model Details page, you can view the operations supported by the model.
- Deploying a Model
- Calling a Model
The operations supported by different models may vary. Check the console for details.

Deploying a Model

You can deploy custom models and fine-tuned models using one of

the following methods. For details, see Deploying a Model Service.

On the Model Square page: Click Model Deployment on a model card.
On the Model Details page: In the upper right corner of the page, click Model Deployment and select a version. Alternatively, click Deploy on the right of the version area.

On the MaaS platform: For details, see Deploying a Model Service.
Figure 1 Deployment on this platform

Figure 2 English GUI

Calling a Model

MaaS provides commercial-use APIs for inference. After subscribing to a model service, you can directly experience or call the model service without waiting for deployment. The model service is billed based on usage (such as the number of tokens or image generation duration).

Use either of the following methods to call a model for inference:
- On the Model Square page: Click Model Deployment on a model card.
- On the Model Details page: In the upper right corner of the page, click Model Deployment and select a version. Alternatively, click Deploy on the right of the version area.
On the View Call Description page, enable or call the model service according to the instructions.
Subscribing to a model service: In the Subscribe to Built-in Services and Try Popular Models Now area, review the information, select I have read and agree to the above terms and MaaS Service Statement, and click Subscribe.

For details about the API key parameters, see Creating an API Key. For details about the API calling parameters, see Model API Calling Specifications.
Figure 3 API calling

Parent Topic: Appendix

Previous topic: Managing API Keys

Next topic: Subscribing to Built-in Services

Feedback

Was this page helpful?

Helpful Not helpful

Provide feedback

Thank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.

The system is busy. Please try again later.

For any further questions, feel free to contact us through the chatbot.

Chatbot