# Build Small Winners Reveal

## Executive summary

The Build Small Hackathon winners reveal celebrated projects that emphasize building highly functional applications using small, efficient models and local/on-device processing. The community demonstrated a strong focus on 'local first' AI solutions, achieving impressive metrics like 64% of apps running fully offline. Winning projects showcased practical utility—such as scam defense (Jawbreaker) or specialized coaching (Posify)—proving that powerful, real-world impact can be achieved with tiny model footprints.

## Key takeaways

- Focus on Local and Small Models: The hackathon emphasized building small apps using sub-32 billion parameter models. Key metrics showed that 64% of submissions ran fully offline, highlighting the viability of 'local first' AI architectures.
- High Utility in Niche Applications: Winning projects demonstrated significant real-world utility. Examples include Jawbreaker (a private scam defense tool for suspicious text/emails) and Agenda Parser (breaking down dense local government agendas), proving the value of small, focused solutions.
- Advanced Technical Implementation: Winners utilized advanced techniques like WebGPU for browser-based real-time games (Parry) and quantization methods (GGUF) to drastically reduce cold boot times, demonstrating high technical polish.
- Open Source Model Ecosystem: The most popular model family used was Qwen (288 spaces), followed by MiniCPM OpenBMB and Nematron, confirming the community's reliance on diverse open-source models.

## Technical details

- Submission Metrics: Out of 946 total spaces, 850 were Gradio Spaces. The deadline rush was notable, with 40% of all submissions arriving in the final 48 hours.
- Model Usage and Efficiency: While the premise required sub-32B parameter models, only 19% actually did so. However, 367 apps used models below 4 billion parameters, and 64% ran fully offline.
- Model Orchestration: One winning app (Viva Mice) was notable for using 21 distinct models in a single application, demonstrating intelligent model selection and orchestration rather than just thin wrappers.
- Frameworks and Tools: Specific tools used included Modal (336 spaces), OpenBMB models (309 spaces), Nvidia models (199 spaces), and OpenAI's Codex (137 spaces) for development assistance.

## Practical implications

- The trend toward local, on-device AI suggests a shift in deployment strategy away from cloud-only APIs, prioritizing privacy and low latency.
- Build engineers should focus on optimizing model quantization (e.g., GGUF) and leveraging browser technologies like WebGPU to achieve smooth, real-time performance for small models.
- The success of niche tools (like local government agenda parsers or scam detectors) validates the market need for highly specific, low-resource AI solutions that solve concrete pain points.

## Topics

Build Engineering, Generative AI, Small Language Models (SLMs), Edge Computing, Open Source ML, Hackathons, StreamYard

Source: https://www.youtube.com/watch?v=tYU7erNVfA0