mradermacher/OpenVLThinker-7B-v1.2-GGUF

Full fine-tune (fewer layers) vs ydeng9/OpenVLThinker-7B-v1.2. mradermacher/OpenVLThinker-7B-v1.2-GGUF matches part of ydeng9/OpenVLThinker-7B-v1.2 (339 of its 729 tensors; the rest of ydeng9/OpenVLThinker-7B-v1.2 has no counterpart). On the shared part: mradermacher/OpenVLThinker-7B-v1.2-GGUF is a GGUF (F16) of different weights. Fine-tune: 93% of the weights changed, across 28 of 28 layers.

Tensors changed: MLP 84/84, Attention 196/196, Norms 52/57, Embeddings 0/1, lm_head 1/1

On the Hub · Downloads on Model Pulse