ZhiPuQingLiu

Overview

Zhipu Qingliu, also known as the BigModel open platform, is Zhipu AI's comprehensive AI development platform built around the GLM family of foundation models. It provides enterprises and developers with a full stack of capabilities ranging from the latest state-of-the-art models to production-ready agent services. The platform is designed to streamline the entire AI application lifecycle, enabling teams to experiment, customize, and deploy intelligent solutions at scale.

At its core, Qingliu hosts a range of flagship models, including the GLM-5.3, which the platform describes as a new flagship foundation model with remarkable enhancements in long-horizon tasks and overall performance comparable to Claude Opus 4.6. Other offerings include GLM-5-Turbo for efficient inference, and multimodal models such as GLM-5V-Turbo and GLM-4.6V that natively process images, video, and text for visual programming tasks. The platform also provides specialized models for image generation (GLM-Image), OCR (GLM-OCR), speech recognition (GLM-ASR), and speech synthesis (GLM-TTS), covering a wide range of AI modalities.

Beyond raw models, Qingliu offers an end-to-end toolchain. The Agent marketplace provides curated agents for diverse industries, directly accessible via API for seamless integration into business flows. MCP services support Model Context Protocol integration, allowing agents to connect with external tools and data. The fine-tuning pipeline enables users to customize GLM models with private datasets, while the knowledge base service supports retrieval-augmented generation (RAG) scenarios. For developers, the platform includes a comprehensive API, documentation, a model playground, and a console for managing resources and deployments. The GLM Coding Plan, a subscription-based coding package, is deeply integrated with mainstream development tools and targets high-frequency coding scenarios.

Qingliu is positioned as a trusted, intelligent AI infrastructure for building real-world applications. It benefits AI teams, startups, and enterprises by reducing the complexity of model selection, integration, and deployment. With a strong emphasis on open APIs and industry scalability, the platform is suited for tasks such as developing intelligent assistants, automating code generation, building multimodal applications, implementing document processing workflows, and deploying domain-specific conversational agents. The platform also offers new-user token packages and competitive pricing to accelerate adoption and experimentation.

Core capabilities

Flagship GLM models

Direct access to GLM-5.3, GLM-5-Turbo, and GLM-5.2 with state-of-the-art performance on long-horizon tasks.

Multimodal foundation models

GLM-5V-Turbo and GLM-4.6V natively handle vision, video, and language inputs for visual programming and agent tasks.

Agent marketplace

Curated, industry-specific agents that can be called directly via API and integrated into production environments.

MCP services

Model Context Protocol support for connecting agents to external tools, data sources, and dynamic workflows.

Fine-tuning pipeline

Customize GLM models on private datasets using the platform's fine-tuning console and APIs.

Knowledge base

Build and manage domain knowledge for retrieval-augmented generation and accurate fact-grounded responses.

GLM Coding Plan

Subscription-based AI coding suite integrated with mainstream development tools for high-frequency coding scenarios.

Speech & audio models

GLM-ASR for real-time transcription with accent adaptation and GLM-TTS for context-driven natural speech synthesis.

Image & OCR capabilities

GLM-Image for text rendering in complex layouts and GLM-OCR for document parsing and structured extraction.

Developer experience

Comprehensive API documentation, model playground, scenario examples, and a unified console for resource management.

Use cases

Developing production-ready conversational agents and intelligent assistants

Building visual programming tools that understand screenshots, images, and code

Implementing retrieval-augmented generation (RAG) with private knowledge bases

Fine-tuning GLM models for industry-specific language tasks

Automating code generation, review, and completion with the GLM Coding Plan

Deploying real-time speech recognition and synthesis for contact centers or IoT

Extracting structured data from documents via GLM-OCR for business workflows