GLM-5.3-GGUF

Overview

GLM-5.3 at Unsloth's Dynamic 4-bit - frontier coding and long-horizon agentic performance (753B total, ~39B active) in a 467 GB llama.cpp download, with the full 1M context window intact.

Good for

• agentic
• chat
• coding

Recommended hardware

≈ 512 GB VRAM to run this 753B model at the UD-Q4_K_XL quantization.

Needs a multi-GPU server node (e.g. 8x 80 GB H100, or 4x 192 GB)

Links

• View on HuggingFace
• Run on Infersec