mikeee/m2v-gemma-embedding-300m

Different architecture vs google/embeddinggemma-300m. Different architecture: only 0 of 314/2 tensors match by name and shape.

On the Hub · Downloads on Model Pulse