GLM vs. Guanaco

Overview

GLM is Z.ai's open-weight model family for reasoning, coding, agents, and long-horizon work. GLM-5.2 is the current generation with a one-million-token context window.

The GLM family began with bilingual pretrained models and evolved through the ChatGLM and GLM-4 generations. GLM-5 scaled the mixture-of-experts architecture to 744B total parameters with 40B active parameters. GLM-5.2, released in June 2026, adds a stable one-million-token context window, stronger long-horizon coding and agent performance, adjustable reasoning effort, and efficiency improvements for sparse attention. Its weights are available under the MIT license.

Initial release: 2022-08-04

Current generation: GLM-5.2

Reference

https://z.ai/blog/glm-5.2

Further Reading

• https://z.ai/blog/glm-5
• https://z.ai/blog/glm-5.2
• https://huggingface.co/zai-org

Overview

Guanaco is an LLM based off the QLoRA 4-bit finetuning method developed by Tim Dettmers et. al. in the UW NLP group. Guanaco achieves 99% ChatGPT performance on the Vicuna benchmark.

Guanaco is an LLM that uses a finetuning method called LoRA that was developed by Tim Dettmers et. al. in the UW NLP group. With QLoRA, it becomes possible to finetune up to a 65B parameter model on a 48GB GPU without loss of performance relative to a 16-bit model. The Guanaco model family outperforms all previously released models on the Vicuna benchmark. However, given the models are based off of the LLaMA model family, commercial use is not permitted.

Initial release: 2023-05-23

Reference

https://github.com/artidoro/qlora

Further Reading

• https://arxiv.org/abs/2305.14314
• https://twitter.com/Tim_Dettmers/status/1661379354507476994