mradermacher/vdr-2b-multi-v1-GGUF
Same weights, converted (fewer layers) vs llamaindex/vdr-2b-multi-v1. mradermacher/vdr-2b-multi-v1-GGUF matches part of llamaindex/vdr-2b-multi-v1 (338 of its 729 tensors; the rest of llamaindex/vdr-2b-multi-v1 has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 0/84, Attention 0/196