mlfoundations-cua-dev/ui-tars-1.5-7b-idm-full-sft-12-frames-videos
Partial fine-tune vs Qwen/Qwen2.5-VL-7B-Instruct. Partial change: 85% of the weights changed (attention, lm_head, mlp, norms), in 28 of 32 layers.
Tensors changed: Embeddings 0/2, Norms 55/122, MLP 84/280, Attention 196/324, lm_head 1/1
1 derivatives checked
- mradermacher/ui-tars-1.5-7b-idm-full-sft-12-frames-videos-GGUF: Same weights, converted (fewer layers)