{
  "id": 1818519,
  "title": "Z.ai's GLM-5.3 with max reasoning scores 60 on the Artificial Analysis Intelligence Index, on par with Kimi K3 but below Opus 5 at 63 and Fable 5 at 62 (@artificialanlys)",
  "url": "https://urgent.news/2026/08/18/z-ais-glm-5-3-with-max-reasoning-scores-60-on-the-artificial-analysis",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-18T22:35:01.000Z",
  "source": {
    "name": "Techmeme",
    "slug": "techmeme",
    "url": "https://x.com/artificialanlys/status/2089830890709135426"
  },
  "original_language": "en",
  "account": null,
  "summary": "Z.ai's GLM-5.3 has achieved a maximum reasoning score of 60 on the Artificial Analysis Intelligence Index. This score puts it on par with Kimi K3, but below Opus 5, which scored 63, and Fable 5, which scored 62, according to Artificial Analysis Intelligence Index.\n\nGLM-5.3's score is an improvement of 7 points from its predecessor, GLM-5.2. Once its weights are released, GLM-5.3 is expected to be tied as the leading open weights model.\n\nOther AI models have also been scored on the Artificial Analysis Intelligence Index. Qwen 3.8 27B scored 52, the same as GPT-5.6 Luna, and one point behind GLM-5.2 and DeepSeek V4 Pro 0813.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}