mradermacher/Llama-3.1-10B-Instruct-GGUF

Same weights, converted vs empirischtech/Llama-3.1-10B-Instruct. Same weights stored in another dtype (BF16 → F16/F32), plus 1 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/83, MLP 0/123, Attention 0/164, lm_head 0/1

On the Hub · Downloads on Model Pulse