mradermacher/Qwen2.5-VL-7B-Instruct-ft-20250814163916929-GGUF

Same weights, converted (fewer layers) vs REEA-GLOBAL/Qwen2.5-VL-7B-Instruct-ft-20250814163916929. mradermacher/Qwen2.5-VL-7B-Instruct-ft-20250814163916929-GGUF matches part of REEA-GLOBAL/Qwen2.5-VL-7B-Instruct-ft-20250814163916929 (339 of its 729 tensors; the rest of REEA-GLOBAL/Qwen2.5-VL-7B-Instruct-ft-20250814163916929 has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/57, MLP 0/84, Attention 0/196, lm_head 0/1

On the Hub · Downloads on Model Pulse