Local LLM model page

GLM 4.6 Air (12B)

Zhipu AI lightweight flagship. Strong bilingual CN/EN with hybrid thinking mode, 200K context and tool calling. Apache 2.0 — excellent alternative to Qwen 3.5 9B on modest GPUs.

Find the best model for my hardwareBrowse all 183 LLMs
Parameters
12B
Minimum RAM
12 GB
Model size
7.5 GB
Quantization
Q4_K_M

Can GLM 4.6 Air (12B) run locally?

GLM 4.6 Air (12B) is best suited for mainstream Macs and PCs with 16 GB RAM. LocalClaw recommends Q4_K_M as the default quantization, with at least 12 GB RAM.

Search term for LM Studio or compatible runtimes: glm-4.6-air

Hugging Face repository: THUDM/GLM-4.6-Air-GGUF

chatcodereasoningstandardgeneral

Strengths

  • Zhipu AI lightweight flagship. Strong bilingual CN/EN with hybrid thinking mode, 200K context and tool calling. Apache 2.0 — excellent alternative to Qwen 3.5 9B on modest GPUs.

Limitations

  • Performance depends heavily on quantization, RAM bandwidth and runtime support.

Best use cases

  • chat
  • code
  • reasoning
  • standard
  • general

Benchmarks

Speed: 8/10

Quality: 8/10

Coding: 8/10

Reasoning: 8/10

Technical details

Developer: glm

License: See model repository

Context window: Unknown tokens

Architecture: See model card

Released: 2026-02