barozp/Qwen3.6-28B-REAP20-A3B-GGUF
Same weights, converted vs 0xSero/Qwen3.6-28B. Same weights stored in another dtype (BF16 → BF16/F32), plus 150 new tensors.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/101, MLP 0/200, Attention 0/40
Same weights, converted vs 0xSero/Qwen3.6-28B. Same weights stored in another dtype (BF16 → BF16/F32), plus 150 new tensors.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/101, MLP 0/200, Attention 0/40