Qwen-Drive-1.0-4B

Tools

• qwen_drive_1_vqa_demo_answer_question Answer a free-form question about a driving-scene image with Qwen-Drive-1.0's VLM. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool. Returns: The model's answer as plain text.
• qwen_drive_1_vqa_demo_detect_boxes Detect 3D objects in a bundled surround-camera frame and render the summary. One forward pass of the shared VLM feeds the BEV head set, which predicts 3D boxes, an occupancy grid and a BEV map segmentation in a single frame. The official renderer plots the predicted boxes projected onto every camera and in the bird's-eye view, next to the recorded ground truth in green. Returns: The rendered summary image (RGB) and a one-line detection summary.
• qwen_drive_1_vqa_demo_plan_scene Plan future ego trajectories for a bundled driving scene. The VLM first writes a one-sentence rationale for the driving decision, then the reward-optimized Planning Expert samples 5 s of future waypoints (10 Hz) from the attention cache that generation produced. Returns: A path to the summary figure, the model's reasoning text, and the, displacement error of the first sample against the recorded human drive.

No tool matches.

As the server's own listing gives them. We never call a tool.