Mixed-precision quantizations of Latitude Games' 12B models, using NVFP4+4/6 for MLP layers and FP8_DYNAMIC for self-attention.
Jesse Mitchell
DataSnake
AI & ML interests
None yet
Recent Activity
updated a collection about 2 hours ago
Roleplaying Quants updated a collection about 2 hours ago
Roleplaying Quants updated a collection about 2 hours ago
Roleplaying QuantsOrganizations
None yet
models 18
DataSnake/CommonCanvas-S-NC_repack
Updated
DataSnake/mitsua-diffusion-one_repack
Updated
DataSnake/CommonCanvas-S-C_repack
Updated
DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-FP8
8B • Updated • 43
DataSnake/Mistral-Nemo-Instruct-2407-Down-4over6
9B • Updated • 11
DataSnake/Mistral-Nemo-Instruct-2407-Down-RTN
9B • Updated • 11
DataSnake/Muse-12B-NVFP4-FP8
Text Generation • 8B • Updated • 45
DataSnake/Wayfarer-2-12B-NVFP4-FP8
Text Generation • 8B • Updated • 18
DataSnake/Wayfarer-12B-NVFP4-FP8
Text Generation • 8B • Updated • 56
DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-FP8-RTN
8B • Updated • 11
datasets 0
None public yet