mradermacher/EraX-VL-2B-V1.5-GGUF
Same weights, converted (fewer layers) vs erax-ai/EraX-VL-2B-V1.5. mradermacher/EraX-VL-2B-V1.5-GGUF matches part of erax-ai/EraX-VL-2B-V1.5 (338 of its 729 tensors; the rest of erax-ai/EraX-VL-2B-V1.5 has no counterpart). On the shared part: Same weights stored in another dtype (F32 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 0/84, Attention 0/196