quimmedes/Qwen3.8-Flash-Next-MTP-GGUF

Update README.md

Update README.md

Update README.md

Update README.md

regenerate MTP: emit output_hc_norm.weight (origin-compatible naming)

regenerate MTP: emit output_hc_norm.weight (origin-compatible naming)

regenerate MTP: emit output_hc_norm.weight (origin-compatible naming)

regenerate MTP: emit output_hc_norm.weight (origin-compatible naming)

upload mtp-Qwen3.8-Flash-Next-Q4_K_M.gguf (post-merge build)

upload mtp-Qwen3.8-Flash-Next-Q6_K.gguf (post-merge build)

upload mtp-Qwen3.8-Flash-Next-Q8_0.gguf (post-merge build)

upload mtp-Qwen3.8-Flash-Next-BF16.gguf (post-merge build)

delete mtp-Qwen3.8-Flash-Next-Q4_K_M.gguf for clean re-upload

delete mtp-Qwen3.8-Flash-Next-Q6_K.gguf for clean re-upload

delete mtp-Qwen3.8-Flash-Next-Q8_0.gguf for clean re-upload

delete mtp-Qwen3.8-Flash-Next-BF16.gguf for clean re-upload

Upload mtp-Qwen3.8-Flash-Next-BF16.gguf with huggingface_hub

upload mtp-Qwen3.8-Flash-Next-Q8_0.gguf

upload mtp-Qwen3.8-Flash-Next-Q6_K.gguf

update mtp-Qwen3.8-Flash-Next-Q4_K_M.gguf with proper hyper-connection mixer weights

docs: update available quantizations and usage instructions

upload mtp-Qwen3.8-Flash-Next-Q4_K_M.gguf

docs: add README with usage instructions and link to cafe-llama.cpp

initial commit