Hugging Face's logo

Collections

Discover the best community collections!

Collections trending this week
Qwen3.8-27B MLX Quantizations
Qwen3.8-27B MLX family: 3-bit VLM base + 3-bit-mtp (RAM-poor, effort=low) + pi 4-bit uniform +MTP (donor format). MTP is oMLX-only.
Qwen3.8-27B MLX Quantizations
Qwen3.8-27B MLX family: 3-bit VLM base + 3-bit-mtp (RAM-poor, effort=low) + pi 4-bit uniform +MTP (donor format). MTP is oMLX-only.