A real 35B MoE (vision, tools, uncensored) in PXQ low-bit quants that run on a $100 Tesla P100 or an 11GB 1080 Ti. Engine: pxq_llama (MIT).
Alex Alvarez PRO
poisonxa
AI & ML interests
None yet
Recent Activity
updated a model about 20 hours ago
poisonxa/Swift-1.5-Qwen3.8-Flash-Next-PXQN2 published a model about 20 hours ago
poisonxa/Swift-1.5-Qwen3.8-Flash-Next-PXQN2 updated a model 1 day ago
poisonxa/Swift-1.5-Qwen3.8-Flash-Next-PXQN-32GBOrganizations
None yet
models 12
poisonxa/Swift-1.5-Qwen3.8-Flash-Next-PXQN-32GB
180B • Updated • 955 • 1
poisonxa/Swift-1.5-Qwen3.8-Flash-Next-PXQN2
180B • Updated • 1
poisonxa/PXA-Supporter-Models
40B • Updated • 276 • 3
poisonxa/PXA-Valued-Models
Updated
poisonxa/PXA-bench-files-GGUF
35B • Updated • 1.53k
poisonxa/Flint-PXQN4
40B • Updated • 143
poisonxa/Swift-1.5-Qwen3.8-27B-PXQN-OneCard
27B • Updated • 2.19k • 2
poisonxa/Qwen3.8-27B-PXQN-OneCard
Text Generation • 27B • Updated • 2.67k • 1
poisonxa/Ornith-1.5-35B-A3B-PXQ4-GGUF
36B • Updated • 782
poisonxa/Qwable-27B-GGUF
27B • Updated • 715 • 1
datasets 0
None public yet