majentik/Qwen3.6-35B-A3B-RotorQuant-GGUF-Q8_0

docs: card hygiene — llama-completion -no-cnv, drop fork-clone quickstart (upstream-first KV guidance)

Remove links to retired duplicate brand variants

Card accuracy sweep: honest brand labeling, remove dead links, upstream KV tip

docs: upstream-first KV-cache guidance (q8_0/q4_0, mainline Hadamard rotation); fork demoted to experimental

docs: remove AWQ variants (AWQ lane paused; MLX focus)

docs: Tier 2 polish — variant matrix + quant trade-off

chore(card): add hardware compatibility section

chore(card): refresh majentik/Qwen3.6-35B-A3B-RotorQuant-GGUF-Q8_0 to 8-section template

docs: initial model card

feat: publish Qwen3.6-35B-A3B GGUF Q8_0

initial commit