It contains 15 small Python repository fixtures, each with a failing implementation and a deterministic acceptance command. The harness sends each task to a real local Ollama model, applies the ...
# python unified_vision_tools_final_rpi5.py --task ocr --image samples/hello.png # python unified_vision_tools_final_rpi7.py --task objects --image samples/objects ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results