mradermacher/InternVL3-38B-Pretrained-GGUF
Quantization (fewer layers) vs OpenGVLab/InternVL3-38B-Pretrained. mradermacher/InternVL3-38B-Pretrained-GGUF matches part of OpenGVLab/InternVL3-38B-Pretrained (771 of its 1366 tensors; the rest of OpenGVLab/InternVL3-38B-Pretrained has no counterpart). On the shared part: mradermacher/InternVL3-38B-Pretrained-GGUF is a GGUF quantization (Q8_0) of the same weights: 321 tensors match exactly and 450 within quantization error.
Tensors changed: Embeddings 0/1, Norms 0/129, MLP 0/192, Attention 0/448, lm_head 0/1