← Back to selected work

Clothing Attribute AI

AI / MOBILE / CASE STUDY

An outfit, decoded in under two seconds.

The problem

Clothing attribute classification needed a fast response in an Android experience.

The approach

Built an Android app backed by FastAPI, using LLaMA 3.2 through Ollama. Quantized the model and shortened prompts to reduce inference latency.

How the system connects

  1. 01Android input
  2. 02FastAPI service
  3. 03Quantized model
  4. 04Clothing attributes

Technology

PythonFastAPILLaMA 3.2OllamaQuantization

Outcome & next steps

Reported inference under two seconds on the tested device setup after model quantization and prompt reduction. Performance depends on hardware and input.

Built at Prishitech Solutions · Sep 2025–present.

Discuss a similar project ↗