GLM Models to Run Locally | Atomic Chat

Reasoning

Code

Multilingual

Web

Vision

Thinking

Tools

Audio

Visuals

Embedding

GLM-5.3-Flash

GLM-5.3-Flash is a 320B MoE with 18B active parameters, the first natively multimodal model in the GLM-5 series, released under MIT.

Thinking

Tools

Vision

Reasoning

Code

GLM-5.1

GLM-5.1 is Z.ai’s 753.9B-parameter MIT-licensed flagship for agentic engineering, topping SWE-Bench Pro at 58.4. Run it locally in Atomic Chat.

Tools

Thinking

Reasoning

Code

GLM-5.2

GLM-5.2 is Z.ai’s 753.3B-parameter flagship with a solid 1M-token context, released under MIT with no regional limits.

Thinking

Reasoning

Code

GLM-4.7-Flash

GLM-4.7-Flash is Z.ai’s 30B-A3B MoE under MIT: 31.2B parameters, about 3B active per token, and 2-bit GGUF builds fit in 12 GB.

Tools

Thinking

Reasoning

Code