bot-lab-21 / GLM-5.3-Int4-Int8Mix-Pollard-GPTQ
Text Generation763B990Updated Sep 8, 2026
Downloads
99
Likes
0
Last updated
Sep 8, 2026
License
glm-5.3-license
A production-serving quantization of zai-org/GLM-5.3 (744B-parameter MoE, ~40B active; 78 layers, 256 routed experts + 1 shared, top-8; MLA attention with DeepSeek-style sparse indexer; 1 MTP layer). Cooked from the native BF16 checkpoint (not from a pre-quantized body) with f…
Tasks & library
- text generation
- vllm
Tags
- vllm
- glm_moe_dsa
- glm
- glm-5.3
- moe
- compressed-tensors
