PromptEnhancer/PromptEnhancer-32B

Full fine-tune vs Qwen/Qwen2.5-VL-32B-Instruct. Fine-tune: 94% of the weights changed, across 64 of 64 layers.

Tensors changed: Attention 448/576, MLP 192/388, Norms 128/194, Embeddings 0/2, lm_head 1/1

2 derivatives checked

• mradermacher/PromptEnhancer-32B-i1-GGUF: Quantization (fewer layers)
• mradermacher/PromptEnhancer-32B-GGUF: Quantization (fewer layers)

On the Hub · Downloads on Model Pulse