mlx-community/Qwen3.6-35B-A3B-8bit
Quantization vs Qwen/Qwen3.6-35B-A3B. mlx-community/Qwen3.6-35B-A3B-8bit is a quantized version of the other (U32/BF16).
15 derivatives checked
- RAS-LAB/Qwen3.6-26B-A3B-KREAP-8bit-MLX: Same weights, repacked
- RAS-LAB/Qwen3.6-30B-A3B-KREAP-8bit-MLX: Same weights, repacked
- neopolita/Qwen3.6-23B-A3B-Niwaki-v2.2-2bit-mlx: Abliterated
- neopolita/Qwen3.6-27B-A3B-Niwaki-2bit-mlx: Same weights, repacked
- neopolita/Qwen3.6-23B-A3B-Niwaki-v2.3-2bit-mlx: Abliterated
- neopolita/Qwen3.6-23B-A3B-Niwaki-v2.3-4bit-mlx: Abliterated
- neopolita/Niwaki-Qwen3.6-24B-A3B-v3.0-mlx: Same weights, repacked
- neopolita/Niwaki-Qwen3.6-30B-A3B-v3.0-mlx: Same weights, repacked
- neopolita/Qwen3.6-11B-A3B-Niwaki-4bit-mlx: Same weights, repacked
- neopolita/Qwen3.6-23B-A3B-Niwaki-v2.1-4bit-mlx: Abliterated
- neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx: Same weights, repacked
- neopolita/Qwen3.6-23B-A3B-Niwaki-v2.1-2bit-mlx: Abliterated
- neopolita/Niwaki-Qwen3.6-27B-A3B-v3.0-mlx: Same weights, repacked
- neopolita/Qwen3.6-19B-A3B-Niwaki-v2-4bit-mlx: Abliterated
- neopolita/Qwen3.6-23B-A3B-Niwaki-v2.2-4bit-mlx: Abliterated
2 other repos hold byte-for-byte identical weight files.