Skip to main content
Original: Simon Willison · 11/02/2026

​
Summary

This is a huge new MIT-licensed model: 754B parameters and 1.51TB on Hugging Face This is a huge new MIT-licensed model: 754B parameters and 1.51TB on Hugging Face twice the size of GLM-4.7 which was 368B and 717GB (4.5 and 4.6 were around that size too).

​
Key Insights

“This is a huge new MIT-licensed model: 754B parameters and 1.51TB on Hugging Face” — Discussing the size and significance of the GLM-5 model.
“It’s interesting to see Z.ai take a position on what we should call professional software engineers building with LLMs” — Commenting on the emerging term ‘Agentic Engineering’ for software development with LLMs.
“I ran my ‘Generate an SVG of a pelican riding a bicycle’ prompt through GLM-5” — Illustrating the practical use of GLM-5 with a creative example.

​
Topics

​
Full Article

# GLM-5: From Vibe Coding to Agentic Engineering
Author: Simon Willison
Published: 2026-02-11
Source: https://simonwillison.net/2026/Feb/11/glm-5/#atom-everything
GLM-5: From Vibe Coding to Agentic Engineering This is a huge new MIT-licensed model: 754B parameters and 1.51TB on Hugging Face twice the size of GLM-4.7 which was 368B and 717GB (4.5 and 4.6 were around that size too). It’s interesting to see Z.ai take a position on what we should call professional software engineers building with LLMs - I’ve seen “Agentic Engineering” show up in a few other places recently. most notable from Andrej Karpathy and Addy Osmani. I ran my “Generate an SVG of a pelican riding a bicycle” prompt through GLM-5 via OpenRouter and got back a very good pelican on a disappointing bicycle frame: The pelican is good and has a well defined beak. The bicycle frame is a wonky red triangle. Nice sun and motion lines.

​
Key Takeaways

​
Notable Quotes

This is a huge new MIT-licensed model: 754B parameters and 1.51TB on Hugging Face
Context: Discussing the size and significance of the GLM-5 model.
It’s interesting to see Z.ai take a position on what we should call professional software engineers building with LLMs
Context: Commenting on the emerging term ‘Agentic Engineering’ for software development with LLMs.
I ran my ‘Generate an SVG of a pelican riding a bicycle’ prompt through GLM-5
Context: Illustrating the practical use of GLM-5 with a creative example.
  • [[topics/agent-native-architecture]]
  • [[topics/prompt-engineering]]
  • [[topics/ai-agents]]

[AINews] Z.ai GLM-5: New SOTA Open Weights LLM

Swyx · reference · 73% similar

Claude Sonnet is a small-brained mechanical squirrel of <T>

Geoffrey Huntley · explanation · 71% similar

ggml.ai joins Hugging Face to ensure the long-term progress of Local AI

Simon Willison · explanation · 71% similar
Originally published at https://simonwillison.net/2026/Feb/11/glm-5/#atom-everything.