Use this page as the fast route into GLM 5.2 API setup. It compares providers, explains OpenRouter and direct Z.ai access, summarizes pricing, clarifies free access, and points coding teams toward the Coding Plan workflow when multi-file changes need more structure.
Developers can use GLM 5.2 through the direct Z.ai API or through OpenRouter with the model ID z-ai/glm-5.2. OpenRouter is usually the fastest path for teams that already use OpenAI-compatible clients, unified billing, or provider routing. Z.ai direct access is the official route and is the better starting point when you want first-party account support and the dedicated coding endpoint.
Compare GLM 5.2 API providersOpenRouter lists z-ai/glm-5.2 at $1.20 per 1M input tokens and $4.10 per 1M output tokens. Direct Z.ai pricing can change by account, region, or product route, so production budgets should be checked against the provider dashboard before launch. The pricing page shows the token math for chat apps, coding agents, and long-context repository review.
See GLM 5.2 pricing examplesDo not plan on unlimited free hosted API usage. Some providers may offer trials, promotional credit, or limited free access, but hosted inference has real serving cost. If you need a free research route, the open weights at zai-org/GLM-5.2 are the more durable path, provided you have enough local or rented compute for your workload.
Understand free vs paid accessCoding Plan is a workflow where GLM 5.2 outlines files, dependencies, and implementation order before writing code. It matters most for multi-file tasks such as API services, migrations, test suites, and agentic coding sessions. The dedicated guide explains how to prompt for a plan, use OpenCode, and decide when planning is worth the extra tokens.
Read the Coding Plan guideAfter the API and pricing questions are clear, use the supporting guides to compare GLM 5.2 against other coding models, check benchmark context, configure OpenCode, or run the open weights locally through supported serving frameworks.