GLM

Releases

1. Aug 15

   GLM-5.3 · Flash

2. Jun 13

   GLM-5.2 · 753B · 40B active · open weights

3. Apr 7

   GLM-5.1 · 744B · 40B active · open weights

Trends

• Chinese labs take the lead in open-weight models
• Open-weight models diversify in architecture, modality, and size
• Frontier labs slow down and publicly report their own risks
• Open weights consolidate around a flash tier and an Nvidia-owned hub

Items

1. Aug 26

   Zhipu's GLM-5.3-Flash

   The GLM-5 series' first natively multimodal model (320B total, 18B active, 1M-token context, hybrid sparse-plus-linear attention) shipped with MIT-licensed weights and runs entirely on domestic Chinese accelerators, after twelve days atop OpenRouter as the anonymous "Ox Alpha."

2. Aug 15

   Z.ai's GLM-5.3 tops open models, weights delayed

   Artificial Analysis scored the post-training-only GLM-5.3 at 60 on its Intelligence Index, tying Kimi K3 as the top open model, but Z.ai delayed the open-weight release about two weeks because its vulnerability-discovery capability needs more safety hardening.

3. Jun 13

   Z.ai's GLM-5.2

   A 753B-parameter MoE (~40B active) with a stable 1M-token context and MIT-licensed weights, scoring 62.1 on SWE-bench Pro versus GPT-5.5's 58.6 and 74.4% on FrontierSWE, with IndexShare attention cutting per-token compute ~2.9x at full context.

4. Apr 7

   Z.ai's GLM-5.1

   A 744B-parameter MIT-licensed MoE (40B active) that tops SWE-Bench Pro at 58.4%, ahead of GPT-5.4, Claude Opus 4.6, and Gemini 3.1 Pro, sustains eight-hour autonomous coding loops, and was trained entirely on Huawei Ascend hardware.