Transparency signal for adopters, based on public metrics — NOT a quality score:
• Adoption: 4.4k GitHub stars (public metric)
• Maintenance: last push ~9 months ago
• Openness: weights + code public (MIT)
• Docs: official HF card, blog, ModelScope
Most of this number is GitHub stars. See /methodology/#scoring.
⚠️ Red Flags & Trade-offs (AMI)
- The flagship GLM-4.5 is 355B/32B — a datacentre model at full precision; only the smaller GLM-4.5-Air (106B/12B) is practical on fewer machines. [Model card]
- Thinking mode adds latency; non-thinking mode is faster but weaker on hard reasoning — pick per task. [GitHub]
- English docs and third-party tooling are thinner than Qwen/DeepSeek; most benchmarks are the authors' own. [GitHub]
Each flag links to its primary source. See our Correction Policy.
📝 Report a correction or complaint
Corrections and complaints go straight to our editorial queue — read and actioned by the site operator, not lost in an inbox. No email needed.
Zhipu AI's 355B/32B open MoE (MIT) that natively fuses reasoning, coding and agent capabilities — open-source SOTA on agentic benchmarks in Jul 2025.
4.4k stars · 486 forks · last push Feb 2026
💡 Lighter Alternatives (If this is too heavy)
Disclaimer
Red Flags and the AMI score are editorial analysis, not endorsements. They are produced with AI assistance and reviewed by a human, but are not a substitute for reading the primary source. The Adoption & Maintainability Index (AMI) reflects public adoption and availability signals only — never a verdict on the science, the code quality, or the people behind a project. We are not affiliated with any lab, university, or company named on this page, and links point only at primary sources (papers, repositories, journals). Content is informational and not engineering, business, or investment advice.