ExLlamav3 QLoRA'ed occult philosophy finetune
UnstableLlama
UnstableLlama
AI & ML interests
Local inference, quantization, training. Philosophy of AI.
Recent Activity
new activity 7 days ago
UnstableLlama/Qwen3.6-35B-A3B-exl3-4.00bpw:Redo with MTP and mul1 quants liked a model 8 days ago
turboderp/IQuest-Coder-V1-40B-Instruct-exl3 updated a collection 9 days ago
Semancer-27BOrganizations
None yet
experimental high-compression abliteration
-
UnstableLlama/Mistral-Medium-3.5-128B-desiccated
125B • Updated • 14 -
mradermacher/Mistral-Medium-3.5-128B-desiccated-i1-GGUF
125B • Updated • 1.41k • 2 -
UnstableLlama/Mistral-Medium-3.5-128B-desiccated-exl3-2.50bpw
21B • Updated • 44 -
mradermacher/Mistral-Medium-3.5-128B-desiccated-GGUF
125B • Updated • 153 • 1
My experimental "desiccated" models
ExLlamav3 QLoRA'ed occult philosophy finetune
Occult Philosophy Finetune
experimental high-compression abliteration
-
UnstableLlama/Mistral-Medium-3.5-128B-desiccated
125B • Updated • 14 -
mradermacher/Mistral-Medium-3.5-128B-desiccated-i1-GGUF
125B • Updated • 1.41k • 2 -
UnstableLlama/Mistral-Medium-3.5-128B-desiccated-exl3-2.50bpw
21B • Updated • 44 -
mradermacher/Mistral-Medium-3.5-128B-desiccated-GGUF
125B • Updated • 153 • 1
sycophantic abliteration
My experimental "desiccated" models
My experimental "dehumidified" models