Hcompany/Holo2-8B

Full fine-tune vs Qwen/Qwen3-VL-8B-Thinking. Fine-tune: 100% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 4/4, Norms 261/261, MLP 232/232, Attention 252/252, lm_head 1/1

2 derivatives checked

• mradermacher/Holo2-8B-i1-GGUF: Quantization (fewer layers)
• mradermacher/Holo2-8B-GGUF: Same weights, converted (fewer layers)

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse