https://www.anchorterminal.com/tools/glm-5-3 Dated changes, score moves with their causes, what our workers noticed, and reviews for GLM-5.3. en Wed, 21 Oct 2026 22:56:18 +0000 https://www.anchorterminal.com/tools/glm-5-3#rev_3735 https://www.anchorterminal.com/tools/glm-5-3#rev_3735 Wed, 21 Oct 2026 00:00:00 +0000 review GLM-5, 5.1, 5.2 and 5.3 each keep their own repository, and Z.ai's blog of 14 August 2026 says what changed from GLM-5.2. The weights followed on 27 August, with template fixes on 4 September and 14 October, the second a community pull request Z.ai merged. No weights were replaced in place, and files and card agree. Replies are thin, with 2 of 16 discussions answered, the last on 8 September. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/glm-5-3#rev_3736 https://www.anchorterminal.com/tools/glm-5-3#rev_3736 Wed, 21 Oct 2026 00:00:00 +0000 review The GLM-5.3 License is MIT-style and free for commercial use, with one condition aimed at the largest operators. A model-as-a-service group above US$10 billion in revenue over 12 months must pass Z.AI's security review first. The FP8 files are 755.7 GB, vLLM's recipe runs them on 8xH200 or 8xH20, and BF16 is about 1.5 TB. Z.ai states no memory needs. Z.ai's API, Ollama's cloud and five providers on Hugging Face serve it. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/glm-5-3#rev_3737 https://www.anchorterminal.com/tools/glm-5-3#rev_3737 Wed, 21 Oct 2026 00:00:00 +0000 review Apart from a benchmark table, GLM-5.3's card holds deployment links and two notes on reasoning_effort and clear_thinking. Tools are defined only in chat_template.jinja as arg_key and arg_value pairs, with results after an observation token, and vLLM and SGLang name the glm47 parser. JSON output is documented for the hosted API alone, and thinking can't be switched off. The 22 independent results span four areas, with 84.2 on Scale SEAL's MCP Atlas and 38.89 on Vals AI's Terminal-Bench 4.0 run. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/glm-5-3#rev_3738 https://www.anchorterminal.com/tools/glm-5-3#rev_3738 Wed, 21 Oct 2026 00:00:00 +0000 review Z.ai's blog says GLM-5.3's release waited on safety evaluation and hardening, yet publishes no results, while the card reports ExploitGym 105 and 130 and ExploitBench 54.4 as capability, Z.ai's own figures. No refusal scope or injection guidance exists. The 141 safetensors shards load through Transformers' glm_moe_dsa code with a SHA-256 each. There's no security.txt or SECURITY.md, and cvd.z.ai lists flaws the model found elsewhere, not a route for reports about it. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/glm-5-3 https://www.anchorterminal.com/tools/glm-5-3#run-2026-10-01 Thu, 01 Oct 2026 00:00:00 +0000 listing GLM-5.3 is Z.ai's open-weight mixture-of-experts model for coding and long-horizon agent work, 744B parameters with 40B active and a 1M-token context, published on Hugging Face in FP8 and BF16 under its own licence.