Multimodal LLMs split into text-only and encoder-only halves, verified against the originals.
-
sartajbhuvaji/unfuse-GLM-4.6-Flash-text
Text Generation • 9B • Updated • 869 • 2 -
sartajbhuvaji/unfuse-GLM-4.6-Flash-text-GGUF
Text Generation • 9B • Updated • 367 -
sartajbhuvaji/unfuse-GLM-4.6V-Flash-mmproj-GGUF
0.9B • Updated • 195 -
sartajbhuvaji/unfuse-Gemma-4-31B-IT-NVFP4-FP8attn-text
Text Generation • 31B • Updated • 48