mradermacher/MiLMMT-46-4B-v0.1-GGUF
Same weights, converted (fewer layers) vs xiaomi-research/MiLMMT-46-4B-v0.1. mradermacher/MiLMMT-46-4B-v0.1-GGUF matches part of xiaomi-research/MiLMMT-46-4B-v0.1 (445 of its 884 tensors; the rest of xiaomi-research/MiLMMT-46-4B-v0.1 has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/205, MLP 0/102, Attention 0/136, lm_head 0/1