OpenGVLab/InternVL3_5-14B
Full fine-tune vs OpenGVLab/InternVL3_5-14B-MPO. Fine-tune: 93% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 0/5, Norms 118/257, MLP 125/222, Attention 160/256, other 0/48, lm_head 1/1
6 derivatives checked
- bartowski/OpenGVLab_InternVL3_5-14B-GGUF: Same weights, converted (fewer layers)
- mradermacher/InternVL3_5-14B-i1-GGUF: Quantization (fewer layers)
- mradermacher/InternVL3_5-14B-GGUF: Quantization (fewer layers)
- lmstudio-community/InternVL3_5-14B-GGUF: Quantization (fewer layers)
- cyankiwi/InternVL3_5-14B-AWQ-4bit: Quantization
- OpenGVLab/InternVL3_5-14B-Flash: Full fine-tune