-
philbert440/Qwen3.8-27B-Uncensored-Aggressive-W4A16-AWQ
Image-Text-to-Text • 27B • Updated • 16.7k • 19 -
philbert440/Qwen3.8-27B-Uncensored-Aggressive-NVFP4
Image-Text-to-Text • 28B • Updated • 536 • 3 -
philbert440/Qwen3.8-27B-Uncensored-Aggressive
Image-Text-to-Text • 27B • Updated • 260 • 6 -
philbert440/Qwen3.8-27B-Uncensored-Cyber
Image-Text-to-Text • 27B • Updated • 2.64k • • 32
Phil Tompkins PRO
philbert440
AI & ML interests
rivethub.io
Recent Activity
liked a model about 14 hours ago
bjivanovich/Qwen3.8-27B-Uncensored-Cyber-ATX-MTP-GGUF new activity about 14 hours ago
philbert440/GLM-5.3-Flash-EXL3:Nice one! updated a model 2 days ago
philbert440/GLM-5.3-Flash-EXL3Organizations
None yet
Uncensored Deckard-40B, Opus-distilled (DavidAU lineage). W4A16-AWQ, V100/SM70. HermesCalibrated adds tool-use / function-calling.
-
philbert440/Qwen3.8-27B-Uncensored-Aggressive-W4A16-AWQ
Image-Text-to-Text • 27B • Updated • 16.7k • 19 -
philbert440/Qwen3.8-27B-Uncensored-Aggressive-NVFP4
Image-Text-to-Text • 28B • Updated • 536 • 3 -
philbert440/Qwen3.8-27B-Uncensored-Aggressive
Image-Text-to-Text • 27B • Updated • 260 • 6 -
philbert440/Qwen3.8-27B-Uncensored-Cyber
Image-Text-to-Text • 27B • Updated • 2.64k • • 32
Heretic abliteration on the ThinkingCap brevity base: faster, equally uncensored. Aggressive-NVFP4 = most open + fastest (70 tok/s, V100).
Uncensored Deckard-40B, Opus-distilled (DavidAU lineage). W4A16-AWQ, V100/SM70. HermesCalibrated adds tool-use / function-calling.