Z.ai API Platform

What it does

Z.ai API Platform is a developer API for the GLM model family. You send text, images, video or files, and get back generated text, reasoning output, structured JSON, images or tool-call results, depending on the model. It covers text models with adjustable reasoning effort and up to 1M tokens of context, vision models, image generation, and built-in web search.

Developers call it over a standard HTTP API, the official Python or Java SDKs, or OpenAI-compatible SDKs, which eases migration from OpenAI. Typical uses are coding assistants, agent backends, and document or image understanding. Function calling, streaming, context caching and structured output are listed features. Coding tools such as Cline, OpenCode, Factory and Crush are cited as users of the GLM models.

It suits developers and teams building AI features who want metered billing across many model sizes, from low-cost Flash tiers to the flagship GLM-5.3. Access requires registering, topping up billing if needed, and creating an API key. It is a programming interface, not a chat app. Custom rate limits, private deployment and consulting go through the sales team, and bundle prices are not shown on the pages reviewed.

What it can do

• Text generation with adjustable reasoning effort and up to 1M-token context
• Function calling to connect models to external tools
• Structured JSON output and streaming responses
• Vision understanding of images, video and files
• Built-in web search and context caching to reduce repeated-context cost
• Image generation with GLM-Image

Works with

• OpenAI Python SDK compatibility
• OpenAI NodeJs SDK compatibility

Example workflows

Add a coding assistant to an app

Input: a code repository excerpt and a task description. Steps: create an API key, call a GLM model through the HTTP API or an SDK, enable function calling for tool use and streaming for live output. Output: generated code or refactoring suggestions.

Extract structured data from documents

Input: images or files plus a JSON schema. Steps: send them to a vision-capable model with structured output enabled, then parse the response. Output: reliable JSON fields for downstream systems.

Before you choose

• Billing is per token, per use or per image depending on the model, so costs vary by model choice and by cached versus uncached input.
• The GLM Coding Plan is a separate monthly subscription that uses a dedicated endpoint, not the pay-as-you-go API.
• Custom rate limits, on-premises deployment and agent development require contacting sales.

Listed under

• Coding
• MCP and agent tools
• Agents
• Developers
• Function calling
• LLM API
• Long context

Get started

Register, top up if needed, create a key

1. Register or log in to the Z.AI Open Platform and top up billing if needed.
2. Create a Standard API Key on the API Keys management page.
3. Install an SDK (for example pip install zai-sdk) or call the chat completions endpoint over HTTP with your key.

Prices per 1M tokens. GLM-5.3: $1.4 input, $0.26 cached input, $4.4 output. GLM-5.3-Flash: $0.15 input, $0.03 cached input, $0.50 output. Web search $0.01 per use. GLM-Image $0.015 per image. Prepaid usage bundles are mentioned, but their prices were not confirmed. Some models are listed as Free without terms.

Source: docs.z.ai/guides/overview/pricing, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Z.ai API Platform — Start building with GLM-5.3 ↗z.ai
• Overview - Overview - Z.AI DEVELOPER DOCUMENT ↗docs.z.ai
• Pricing - Overview - Z.AI DEVELOPER DOCUMENT ↗docs.z.ai
• Quick Start - Overview - Z.AI DEVELOPER DOCUMENT ↗docs.z.ai

Company website

docs.z.ai: product, pricing and policy pages

YC directory mirror

Batch and company name

Human review

Checked by a person

knowledge_base

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.