mradermacher/UI-TARS-1.5-7B-i1-GGUF
Quantization (fewer layers) vs ByteDance-Seed/UI-TARS-1.5-7B. mradermacher/UI-TARS-1.5-7B-i1-GGUF matches part of ByteDance-Seed/UI-TARS-1.5-7B (339 of its 729 tensors; the rest of ByteDance-Seed/UI-TARS-1.5-7B has no counterpart). On the shared part: mradermacher/UI-TARS-1.5-7B-i1-GGUF is a GGUF quantization (Q6_K) of the same weights: 141 tensors match exactly and 198 within quantization error.
Tensors changed: Embeddings 0/1, Attention 0/196, Norms 0/57, MLP 0/84, lm_head 0/1