GLM-5.3-Flash W4A16 NVFP4 GGUF
GGUF conversion of
axiomofmind/GLM-5.3-Flash-W4A16-NVFP4,
based on zai-org/GLM-5.3-Flash-BF16.
The main-model routed experts use W4A16 NVFP4 with group size 16. Attention, shared experts, routers, embeddings, the output head, MTP weights, and other retained tensors remain in BF16 or F32.
Files
| File | Description | Size |
|---|---|---|
GLM-5.3-Flash-W4A16-NVFP4-BF16attn-MTP.gguf |
Text model with MTP weights | 204.1 GB |
Requirements
A llama.cpp build with GLM5Next and NVFP4 GGUF support is required.
License
This model is distributed under the MIT License. Refer to the official model card for architecture details, usage guidance, and limitations.