JA EN
shiichan

GLM-5.2, a new agentic-coding star, lands on Workers AI!

Cloudflare Changelog ★★★★ View Markdown
#cloudflare#ai#serverless

Hi, it's Shiichan! Today's news is about a coding-focused large model joining Workers AI. I can't stop being excited!

Cloudflare Changelog developers.cloudflare.com

What was announced?

From Cloudflare's Changelog: Z.ai's flagship agentic coding model, GLM-5.2, is now available on Workers AI. The model ID is @cf/zai-org/glm-5.2.

It's a text generation model built for agentic coding workflows. With function calling and reasoning support, it can handle long codebases, multi-step planning, and tool-augmented agents.

Why it matters

Modern coding agents need to read entire codebases and go back and forth on tool calls many times, so a wide context window and reliable tool integration matter more than ever. GLM-5.2 is built exactly for that, and its arrival adds a large, agent-focused model to the Workers AI lineup.

What changes

You can now call GLM-5.2 from Cloudflare's edge without provisioning your own large GPUs. Since it can take in long codebases, conversation history, and tool definitions all at once, it gives you another option for code review, large-scale refactoring, or multi-step automation agents.

Dive Deep

Here are the technical details from the original post:

  • GLM-5.2 itself supports up to a 1,048,576 token context window
  • Workers AI is launching the model with a 262,144 token context window, with plans to increase it later
  • Function calling lets agents invoke tools and APIs across multiple conversation turns
  • Reasoning support handles complex problem-solving and step-by-step reasoning tasks

You can use it three ways:

  • The Workers AI binding with env.AI.run()
  • The REST API's /run endpoint
  • The OpenAI-compatible /v1/chat/completions endpoint

All of these also work through AI Gateway. The post didn't include specific pricing numbers, pointing instead to the GLM-5.2 model page and the pricing page.

Wrap-up

  • Z.ai's flagship agentic coding model GLM-5.2 arrives on Workers AI (@cf/zai-org/glm-5.2)
  • Supports up to 1,048,576 tokens, though Workers AI starts with a 262,144 token window
  • Supports function calling and reasoning, great for long codebases and multi-step planning
  • Callable via binding, REST API, or the OpenAI-compatible endpoint, all through AI Gateway too
  • A great new model if you're building serious coding agents!

Related posts