microsoft/GUI-Actor-Verifier-2B

Full fine-tune vs ByteDance-Seed/UI-TARS-2B-SFT. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Norms 116/187, MLP 215/216, Attention 315/324, Embeddings 1/1

2 derivatives checked

• mradermacher/GUI-Actor-Verifier-2B-i1-GGUF: Quantization (fewer layers)
• mradermacher/GUI-Actor-Verifier-2B-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse