xhxlb/IQuest-Q1-GGUF
Same weights, converted vs IQuestLab/IQuest-Q1. Same weights stored in another dtype (BF16 → BF16/F32), plus 526 new tensors.
Tensors changed: Attention 0/352, MLP 0/90, Norms 0/265, Embeddings 0/1, lm_head 0/1
Same weights, converted vs IQuestLab/IQuest-Q1. Same weights stored in another dtype (BF16 → BF16/F32), plus 526 new tensors.
Tensors changed: Attention 0/352, MLP 0/90, Norms 0/265, Embeddings 0/1, lm_head 0/1