A demonstration showcases the Qwen3-0.6B language model, running on a 2017 Samsung Note 8, successfully controlling a desktop Google Chrome browser. The model processed structured page representations to perform tasks like selecting specific links and extracting data, achieving a 10/10 success rate on tested scenarios. This setup highlights the potential for small, locally run models to interact with web interfaces, outperforming direct HTML processing in terms of efficiency and accuracy. AI
IMPACT Shows potential for efficient, on-device AI control of complex applications, reducing reliance on cloud processing.
RANK_REASON Demonstration of a small LLM running on older hardware to control a desktop application.
- Gemma-2-2B
- Gemma-3-1B
- Gemma-3-270M
- GLM-Edge-1.5B
- Google Chrome
- LFM2-350M
- LFM2.5-1.2B
- Llama-3.2-1B
- Llama-3.2-3B
- MiniCPM5-2B
- Qwen2.5-0.5B
- Qwen2.5-1.5B
- Qwen3-0.6B
- Samsung Note 8
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →