I did already move them to the QuantPasture. And while I appreciate the kind words, I don't really do dense model quants. The style of MoE-optimized quant that I do (where I keep most of the model in high quality and focus the quantization efforts of the sparsely activated experts) doesn't transfer to non-MoE quants like Qwen3.8-27B. Any quants I made there would basically be the same as @bartowski 's, so I'd recommend going with his!
Aes Sedai PRO
AesSedai
AI & ML interests
None yet
Recent Activity
new activity 1 day ago
AesSedai/MiMo-V2.6-Flash-GGUF:Pro? new activity 1 day ago
AesSedai/MiMo-V2.6-Flash-GGUF:Why not GSK format? updated a dataset 1 day ago
AesSedai/reference-logits