schneewolflabs/Compactor-Qwen3.5-4B
Same weights, repacked (fewer layers) vs Lazarus-Ai/ReAligned-Qwen3.5-4B. schneewolflabs/Compactor-Qwen3.5-4B matches part of Lazarus-Ai/ReAligned-Qwen3.5-4B (426 of its 723 tensors; the rest of Lazarus-Ai/ReAligned-Qwen3.5-4B 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 0/105, Attention 0/224, MLP 0/96
1 derivatives checked
- mradermacher/Compactor-Qwen3.5-4B-GGUF: Same weights, converted