xlangai/Jedi-7B-1080p
Partial fine-tune vs Qwen/Qwen2.5-VL-7B-Instruct. Partial change: 86% of the weights changed (attention, lm_head, mlp, norms), in 28 of 32 layers.
Tensors changed: Embeddings 0/2, Norms 58/122, MLP 88/280, Attention 196/324, lm_head 1/1
3 derivatives checked
- bartowski/xlangai_Jedi-7B-1080p-GGUF: Same weights, converted (fewer layers)
- mradermacher/Jedi-7B-1080p-GGUF: Same weights, converted (fewer layers)
- lmstudio-community/Jedi-7B-1080p-GGUF: Quantization (fewer layers)