Repoint build instructions to charlie12345/ROCmFPX (ROCmFPX FP3/4/6/8 repo)

Upload README.md with huggingface_hub

consolidate to single best-balance rocmfp4 quant

consolidate to single best-balance rocmfp4 quant

Card: best-of-both recommended + KL findings + bundled mmproj/vision (MoE is multimodal)

Add mmproj-F32.gguf (best-of-both / vision)

Add Qwen3.6-35B-A3B-MTP-ROCmFP4-COHERENT-embF16-headQ6.gguf (best-of-both / vision)

Add Qwen ASCII logo header (centered logo+title, full-width spec table)

Claim audit: remove cross-model generalizations (Q6-head %, f16-draft-head, decode comparisons) — state per-model config only, no borrowed sibling measurements

Card: datasheet/neo-brutalist restyle (monospace, hard borders, spec grid, numbered sections); claims audited (f16 KV stated as config only)

Add froggeric v20 chat template (referenced by launch cmd)

provenance: set base_model_relation=quantized

Card: document Q6-head variant (honest: quality measured on 27B) + HF badge note

Add Q6-head variant (output.weight -> Q6_K); quality measured on 27B sibling

Upload Qwen3.6-35B-A3B-MTP-ROCmFP4-STRIX-embF16.gguf with huggingface_hub

Upload README.md with huggingface_hub

initial commit