Sakalti/SabaVL1-2B

Same weights, converted vs unsloth/Qwen2-VL-2B-Instruct. Same weights stored in another dtype (BF16 → F16).

Tensors changed: Embeddings 0/2, Norms 0/187, MLP 0/216, Attention 0/324

1 derivatives checked

• mradermacher/SabaVL1-2B-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse