aisingapore/Qwen-SEA-LION-v4-8B-VL
Full fine-tune vs Qwen/Qwen3-VL-8B-Instruct. Fine-tune: 92% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 3/4, Norms 230/261, MLP 201/232, Attention 227/252, lm_head 1/1
4 derivatives checked
- aisingapore/Qwen-SEA-Guard-8B-2602: Partial fine-tune
- mradermacher/Qwen-SEA-LION-v4-8B-VL-i1-GGUF: Quantization (fewer layers)
- mradermacher/Qwen-SEA-LION-v4-8B-VL-GGUF: Same weights, converted (fewer layers)
- MagicalAlchemist/Qwen-SEA-LION-v4-8B-VL-Magic_decensored: Abliterated