inclusionAI/Ming-flash-omni-Preview

Full fine-tune vs inclusionAI/Ling-flash-2.0. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: MLP 1793/1793, Embeddings 1/1, Attention 64/64, Norms 65/65, lm_head 1/1

2 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse