mradermacher/Falcon-H1-7B-Instruct-GGUF

Same weights, converted vs tiiuae/Falcon-H1-7B-Instruct. Same weights stored in another dtype (BF16 → F16/F32), plus 44 new tensors.

Tensors changed: Embeddings 0/1, MLP 0/132, Norms 0/89, Attention 0/176, lm_head 0/1

On the Hub · Downloads on Model Pulse