---
title: "GLM-5.2 (Fully Tested): I got EARLY ACCESS & This MODEL is CRAZY!"
url: https://daily.dev/posts/glm-5-2-fully-tested-i-got-early-access-this-model-is-crazy--ybo2miby0
source_url: https://www.youtube.com/watch?v=MkFThJWJgg8
type: video:youtube
source: "AICodeKing"
published: 2026-06-13T10:59:14.520Z
updated: 2026-06-17T15:50:33.028Z
reading_time: 6
upvotes: 0
comments: 0
language: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# GLM-5.2 (Fully Tested): I got EARLY ACCESS & This MODEL is CRAZY!

**[AICodeKing](https://daily.dev/sources/aicodeking)** · 6 min read · 0 upvotes · 0 comments

## Summary

GLM 5.2 is an upcoming AI coding model from ZAI, reviewed here with early access. Key features include a 1 million token context window, open weights (likely MIT licensed), and availability on the GLM coding plan ($8/month). Benchmarked against tasks like elevator simulators, 3D modeling with Three.js, SVG generation, and a bow-and-arrow game, it scored 81.43 on the reviewer's custom benchmark — about 6% below Claude Opus 4.8 and Fable. The reviewer recommends pairing it with OpenCode and highlights its speed improvements, focused token usage, and strong price-to-performance ratio compared to Codex or Claude subscriptions.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.youtube.com/watch?v=MkFThJWJgg8>

---

[View this post on daily.dev](https://daily.dev/posts/glm-5-2-fully-tested-i-got-early-access-this-model-is-crazy--ybo2miby0)