🧩 Qwen Runner Vision

Deterministic 12-task extraction + fusion — stick an image in, get the full JSON readout (task JSONs + FusedScene + fused prompt). Specialists run on ZeroGPU; fusion is CPU-only. Every option below is a live toggle.

Image

Tasks to show (all 12 always computed)

image_classification bbox_grounding ocr_text data_type_differentiation data_type_utilization structural_spatial_awareness depth_analysis subject_fixation segmentation outline_association style_structural_awareness semantic_association

VLM/DEFER (no deterministic builder)

vit_accuracy_to_prompt geometric_3d_object_id camera_rotational_offset

Captions (one per line — enrich fusion)

Detections · masks · subject · outline

Depth (near → far)

Fused prompt

Fusion confidence

Download row (JSON)

tasks_json (12 tasks)

tasks_valid

caption structs

fused_json (FusedScene)

age_audit

timing

Examples

·

Built with Gradio logo

·