mlx-community/Mistral-Small-3.2-24B-Instruct-2506-bf16
Same weights, repacked (fewer layers) vs mistralai/Mistral-Small-3.2-24B-Instruct-2506. mlx-community/Mistral-Small-3.2-24B-Instruct-2506-bf16 matches part of mistralai/Mistral-Small-3.2-24B-Instruct-2506 (363 of its 585 tensors; the rest of mistralai/Mistral-Small-3.2-24B-Instruct-2506 has no counterpart). On the shared part: Same weights in different files (re-saved, re-sharded or new metadata).
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 0/120, Attention 0/160, lm_head 0/1
1 derivatives checked
- mlx-community/Mistral-Small-3.2-24B-Instruct-2506-4bit: Quantization