Model Detail

GLM-5.3-Flash-FP4-Local is a multimodal model with 87.6B parameters released by NVIDIA. The model is registered under the image-text-to-text pipeline tag on Hugging Face, distributed under a other license.

GLM-5.3-Flash-FP4-Local ships with 87.6B parameters. Total weight footprint is approximately 175.2 GB, which is the relevant figure when planning local-inference VRAM. Distribution is governed by the other license — review the exact terms before commercial deployment.

GLM-5.3-Flash-FP4-Local is best fit for mixed text-and-image reasoning tasks such as document understanding. Treat this as a starting matrix rather than a benchmark verdict — the right deployment usually depends on the specific evaluation suite that mirrors your workload.