GLM-5.2 Takes the Top Spot Among Text-Only Open-Weights LLMs
Original: GLM-5.2 is probably the most powerful text-only open weights LLM
Z.ai's 753B MIT-licensed GLM-5.2 leads Artificial Analysis's Intelligence Index and ranks second on Code Arena WebDev.
Z.ai has released GLM-5.2, a 753B-parameter MIT-licensed open-weights model with a 1-million-token context window. Independent benchmark site Artificial Analysis ranks it first among open-weights models on their Intelligence Index v4.1, ahead of MiniMax-M3, DeepSeek V4 Pro, and Kimi K2.6. It also places second on Code Arena's WebDev leaderboard behind only Claude Fable 5, despite being text-only, and is available on OpenRouter at $1.40/$4.40 per million input/output tokens.
Chinese AI lab Z.ai released GLM-5.2 in two stages: first to coding-plan subscribers on June 13, 2026, then as full open weights under an MIT license on June 16, 2026. The model is a 753-billion-parameter Mixture-of-Experts (MoE) architecture with 40 billion active parameters and weighs 1.51 TB. It is a text-input-only model — Z.ai maintains a separate vision-capable family (most recently GLM-5V-Turbo), but that line remains closed-weights. The headline upgrade over its predecessor GLM-5.1 is a dramatically expanded context window: 1 million tokens, up from 200,000.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Simon Willison's Weblog →Related
Summaries are AI-generated; the original article is authoritative.