Simon Willison · 2026-06-17 · notable
Simon Willison: GLM-5.2 is probably the most powerful text-only open weights LLM
Simon Willison calls Z.ai's GLM-5.2 today's strongest open-weights text LLM: top of Artificial Analysis Intelligence Index v4.1 at 51, second on Code Arena WebDev behind Claude Fable 5, and ~$1.40/$4.40 per 1M tokens on OpenRouter vs GPT-5.5's $5/$30.
Simon Willison ranks GLM-5.2 as today's top open-weights text LLM — frontier-class scores at roughly a quarter of GPT-5.5's price.
What is it?
Simon Willison's June 17 post argues that Z.ai's recently released GLM-5.2 — a 753B-parameter Mixture-of-Experts model under MIT license with a 1M-token context window — is currently the strongest text-only open-weights LLM. Willison points to Artificial Analysis Intelligence Index v4.1, where GLM-5.2 ranks first among open weights with a score of 51, and to Code Arena's WebDev leaderboard, where GLM-5.2 sits second behind only Claude Fable 5.
How does it work?
Willison runs GLM-5.2 through OpenRouter at roughly $1.40 per million input tokens and $4.40 per million output tokens — meaningfully cheaper than GPT-5.5 ($5/$30) or Claude Opus 4.5–4.8 ($5/$25). He notes one tradeoff flagged by Artificial Analysis: GLM-5.2 uses about 43k output tokens per task versus 26k for GLM-5.1, which dents the cost advantage on long agent runs. Willison also reruns his SVG benchmarks — the pelican-on-bicycle improves over GLM-5.1, but the opossum-on-e-scooter regresses.
Why does it matter?
If you can run a 1.51 TB checkpoint or are happy on OpenRouter, GLM-5.2 closes most of the gap to the closed frontier on coding and general intelligence at a quarter to a sixth of the price. Simon Willison's framing matters because it sets the open-weights conversation: GLM-5.2 is now the default candidate for 'self-host or cheap API.' The increased token spend is the catch buyers should price in.
Who is it for?
AI engineers comparing open-weights models, OpenRouter and self-host users, cost-sensitive teams