OpenGVLab/InternVL3_5-241B-A28B

Partial fine-tune vs OpenGVLab/InternVL3_5-241B-A28B-MPO. Partial change: 71% of the weights changed (attention, mlp, norms), in 94 of 94 layers.

Tensors changed: Norms 252/557, MLP 1650/1836, Attention 376/511, lm_head 0/1, Embeddings 0/5, other 0/90

1 derivatives checked

• OpenGVLab/InternVL3_5-241B-A28B-Flash: Full fine-tune

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

On the Hub · Downloads on Model Pulse