The GLM-5 large language model, developed by Z.ai, is now available for free local execution via Ollama. This Mixture-of-Experts model boasts 744 billion total parameters with 40 billion active per token, trained on 28.5 trillion tokens. It supports a context window of approximately 198K tokens using DeepSeek Sparse Attention and achieves competitive benchmark scores in areas like mathematical reasoning and coding. Ollama simplifies the process of downloading, quantizing, and serving GLM-5 on local hardware, offering an OpenAI-compatible API for seamless integration with existing tools and code. AI
IMPACT Enables local execution of a high-performance LLM, reducing cloud costs and data privacy concerns for developers.
RANK_REASON Frontier-lab model release with system card [lever_c_demoted from frontier_release: ic=1 ai=1.0]
- AIME 2026 I
- Apidog
- DeepSeek Sparse Attention
- GLM-5
- GPQA Diamond
- Hugging Face
- ModelScope
- Ollama
- SWE-bench Verified
- Z.ai
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →