prism-ml/Ternary-Bonsai-2-27B-mlx-2bit
Quantization vs Qwen/Qwen3.8-27B. prism-ml/Ternary-Bonsai-2-27B-mlx-2bit is a quantized version of the other (U32/F16).
8 derivatives checked
- dealignai/Bonsai-2-27B-CRACK-Ternary-JANG: Partial fine-tune
- dealignai/Bonsai-2-27B-CRACK-1.75bit-JANG: Partial fine-tune
- KridgeDookie/Ternary-Bonsai-2-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS-MLX-Mixed-2-8bit: Same weights, repacked
- KridgeDookie/Ternary-Bonsai-2-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS-MLX-Mixed-2-4bit: Same weights, repacked
- OsaurusAI/Bonsai-2-27B-Ternary-JANG: Partial fine-tune
- EigenLabs/Ternary-Bonsai-2-27B-MLX-2bit: Identical copy
- OsaurusAI/Bonsai-2-27B-1.75bit-JANG: Partial fine-tune
- inductiveML/Ternary-Bonsai-2-27B-mlx-lossless-1.75bpw: Same weights, repacked
5 other repos hold byte-for-byte identical weight files.