This model was converted to openvino IR format using the optimum-cli command line tool.

  • Language model weights are in int4
  • input/output embeddings on language model are int8
  • vision encoder in int8
  • calibration dataset applied during vision encoder quantization using optimum-intel defaults.
Downloads last month
29
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Echo9Zulu/Qwen3-VL-4B-Instruct-int4_asym-ov

Finetuned
(240)
this model