bartowski/granite-embedding-107m-multilingual-GGUF

Same weights, converted vs ibm-granite/granite-embedding-107m-multilingual. Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/4, Norms 0/24, Attention 0/48, other 0/24

On the Hub · Downloads on Model Pulse