octopusmegalopod/some-paddleocr1.5-vl-ggufs
Quantization (fewer layers) vs PaddlePaddle/PaddleOCR-VL-1.5. octopusmegalopod/some-paddleocr1.5-vl-ggufs matches part of PaddlePaddle/PaddleOCR-VL-1.5 (165 of its 620 tensors; the rest of PaddlePaddle/PaddleOCR-VL-1.5 has no counterpart). On the shared part: octopusmegalopod/some-paddleocr1.5-vl-ggufs is a GGUF quantization (Q8_0) of the same weights: 37 tensors match exactly and 128 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/37, MLP 0/54, Attention 0/72