lmstudio-community/InternVL3_5-14B-GGUF
Quantization (fewer layers) vs OpenGVLab/InternVL3_5-14B. lmstudio-community/InternVL3_5-14B-GGUF matches part of OpenGVLab/InternVL3_5-14B (443 of its 789 tensors; the rest of OpenGVLab/InternVL3_5-14B has no counterpart). On the shared part: lmstudio-community/InternVL3_5-14B-GGUF is a GGUF quantization (Q8_0) of the same weights: 161 tensors match exactly and 282 within quantization error.
Tensors changed: Attention 0/160, Norms 0/161, MLP 0/120, Embeddings 0/1, lm_head 0/1