358.34B MoE scale
Safetensors metadata reports 358337791296 parameters, described in the card as 358B params with BF16 and F32 tensor types.
Open Source Model Profile · zai-org
GLM-4.7
GLM-4.7 is a 358.34B-parameter glm4_moe text-generation model from zai-org. Its model card documents vLLM and SGLang local deployment with thinking-mode parameters.
- Publisher
- zai-org
- Task
- text-generation
- Model type
- glm4_moe
- License
- mit
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
GLM-4.7 is published by zai-org as a conversational text-generation model. The captured configuration identifies Glm4MoeForCausalLM with a glm4_moe model type, and Safetensors metadata reports 358337791296 parameters. The model card points to Z.ai API Platform services and documents local deployment through vLLM and SGLang.
Recorded capabilities
vLLM and SGLang deployment
According to the model card, local deployment supports vLLM and SGLang, currently on their main branches with official Docker images.
Thinking-mode parameters
The model card documents Preserved Thinking mode and chat_template settings for enabling or disabling thinking in vLLM and SGLang.
Documented evaluation settings
According to the model card, default evaluation uses temperature 1.0, top-p 0.95, and up to 131072 new tokens, with separate settings for Terminal Bench and related tasks.
Use cases in the source record
- Conversational text-generation workflows consistent with the captured conversational and text-generation tags.
- Local deployment experiments using the card's documented vLLM and SGLang instructions, including the thinking-mode settings.
Limitations and unknowns
- No benchmark scores were extracted from this record; only evaluation parameter settings are documented in the card.
- No context-window value was extracted from this record.
- Provider state is historical snapshot data, not independently refreshed current availability.
- Deployment and API details come from the publisher model card and have not been independently verified by Ethen.
Source and provenance
Source: zai-org/GLM-4.7
Captured: Unknown. Processed: 2026-09-07T19:35:01.535249+00:00.
GLM-4.7 👋 Join our Discord community. 📖 Check out the GLM-4.7 technical blog , technical report(GLM-4.5) . 📍 Use GLM-4.7 API services on Z.ai API Platform. 👉 One click to GLM-4.7 . Introduction GLM-4.7 , your new coding partner, is coming with the following features: Core Coding : GLM-4.7 brings clear gains, compared to its predecessor GLM-4.6, in multilingual agentic coding and terminal-based tasks, including (73.8%, +5.8%) on SWE-bench, (66.7%, +12.9%) on SWE-bench Multilingual, and (41%, +16.5%) on Terminal Bench 2.0. GLM-4.7 also supports thinking before acting, with significant improvements on complex tasks in mainstream ag…
F001F002F003F004F005F006F007F008F009F010F013F015F016F017F018F019F020