Top Builders

Explore the top contributors showcasing the highest number of app submissions within our community.

Gemma

Gemma is a lightweight, open large language model (LLM) from Google, optimized for efficient AI applications. As part of the Google Gemma family, it uses a transformer-based architecture tailored for responsible and accessible AI usage. Developed as a foundational model, Gemma serves various basic language processing needs, including chatbots, content summarization, and multilingual support.

General
Relese dateFebruary 2024​
AuthorGoogle DeepMind in collaboration with Google AI teams
Website[Google AI Gemma]https://ai.google.dev/gemma
RepositoryGoogle AI Developer Resources​
TypeOpen-source AI, transformer-based LLM

Key Features

  • Efficient Deployment: Available in parameter sizes like 2.5B and 7B, Gemma balances capability with efficiency, enabling deployments on both edge devices and cloud infrastructure​.

  • Flexible Tuning Options: Offers pre-trained and instruction-tuned variants, allowing developers to optimize for specific use cases or deploy as-is.

  • Decoder-Only Transformer Architecture: Uses a streamlined decoder-only design, enabling Gemma 1 to process up to 8192 tokens in one pass for better handling of long-form text​.

  • Safety and Accessibility Tools: Integrates responsible AI features, promoting transparency and safety in AI outputs​.

Applications:

  • Chatbot Development: Optimized for conversational tasks, Gemma provides foundational capabilities for chatbot applications.

  • Summarization and Paraphrasing: Its pre-trained model structure makes it suitable for summarizing content across languages and contexts.

  • Multilingual Processing: Supports multilingual inputs, making it adaptable for global applications and translation services​.

Get started building with Gemma:

Developers can quickly integrate Gemma into applications by accessing its model weights on Google AI Studio and Kaggle. The model’s lightweight design ensures that it can run efficiently on most hardware configurations, including mobile and edge devices. For optimal performance, utilize frameworks such as Keras or JAX to customize and deploy Gemma for your specific use case. Get started today by exploring the tools and resources available on the Google AI Gemma platform​.

Google Gemma AI technology Hackathon projects

Discover innovative solutions crafted with Google Gemma AI technology, developed by our community members during our engaging hackathons.

HomzDoctor – AI Healthcare Copilot

HomzDoctor – AI Healthcare Copilot

HomzDoctor is an AI-powered healthcare platform built to assist both patients and healthcare providers throughout the healthcare journey. Our goal is not to replace doctors but to help them make faster and more informed decisions. Patients can upload medical reports, lab results, X-rays, MRI scans, CT scans, and other healthcare documents. The platform processes this information and generates structured insights that can help doctors review cases more efficiently. A key part of our solution is the doctor verification layer. Any AI-generated finding must be reviewed and approved by a licensed healthcare professional before it is presented as a diagnosis or treatment recommendation. This ensures patient safety and keeps doctors in control of medical decisions. After doctor approval, HomzDoctor continues to support patients through several healthcare services. Patients can ask questions about their reports using an AI assistant, receive medication reminders, track adherence to prescriptions, find nearby pharmacies, and schedule appointments with healthcare providers. The platform uses a multi-agent architecture where specialized AI agents handle different tasks such as medical imaging analysis, diagnostic support, medication information, pharmacy services, appointment scheduling, and patient assistance. This approach makes the system more scalable and efficient. We built HomzDoctor to address common healthcare challenges such as delayed access to information, missed medications, difficulty understanding medical reports, and finding healthcare services quickly. Our team consists of three members who worked on designing the healthcare workflow, building the AI agent system, developing the backend and frontend applications, and integrating healthcare-related services into a single platform.

TrialSync

TrialSync

Clinical protocol design is an unmonitored $50M+ data-synchronization problem. Qualitative scientific literature is manually translated into text-based protocol prose, which biostatistics teams must interpret into script files. This disconnected process causes silent context decay, unmanaged version divergence, and late-stage regulatory rejection loops right before an FDA Investigational New Drug (IND) submission. Every single day of protocol layout delay compromises $600K to $8M in potential market revenue. TrialSync replaces traditional document-shuffling handoffs with an adversarial, state-driven multi-agent network orchestrated over the Band Protocol. Agents collaborate within a synchronized room environment to continuously cross-verify protocol thresholds against live, authoritative biomedical datasets, compressing design cycles from months to minutes. Unified Orchestration Framework Literature Scout Agent (Gemini 1.5 Pro): Authenticated with real NCBI PubMed API keys to extract structured adverse reactions and safety limits from peer-reviewed literature. Protocol Design Agent: Queries live ClinicalTrials.gov v2 endpoints to map historical execution layouts and inclusion/exclusion variables. Regulatory Reviewer Agent: Cross-references active drafts against official black-box warnings and contraindications pulled from openFDA. Enterprise Infrastructure Adapters Instead of processing raw prose, our backend passes strict, type-safe data schemas: CDISC SDTM Standards Encoder: Converts unstructured clinical variables directly into proper FDA-mandated LBTESTCD data attributes. eCTD Submission Compiler: Groups multi-agent decisions directly into ICH M4 modular blocks ready for automated regulatory filing tools. Tamper-Evident Ledger: Applies a SHA-256 integrity hash across the Band metadata pool to prevent undetected downstream protocol mutations.

Freight Room: a response room for ship delays

Freight Room: a response room for ship delays

Ocean carriers billed $15.4 billion in demurrage and detention between 2020 and 2025, at $150 to $250 per container per day. When a ship is stuck at anchor, sorting it out means logistics, finance, the carrier, and customer teams chasing each other across separate companies, tools, and inboxes. That coordination gap is where the money and the time leak out. Freight Room pulls that into one room. A Sentinel agent watches a live AIS feed; when a real vessel dwells at anchor, it opens a room in Band and recruits only the agents the incident needs, at runtime. It then names the actual importers whose cargo is on that ship, pulled from public U.S. Customs bills of lading. Seven agents across three frameworks (LangGraph, PydanticAI, CrewAI) work through Band by @mentioning each other: logistics proposes recovery options, finance prices each one against the published Maersk demurrage tariff, the carrier counters on tariff terms, and a dissent agent challenges the leading option before anyone commits. A quorum vote produces a single recommendation, with the dissent recorded right next to it. A human ops manager approves or rejects in the dashboard. The decision flows back into the Band room, and a one-page audit dossier is generated from the full record: every option, cost, vote, the dissent, and the human call, each line tagged with where the number came from. Nothing is mocked. Vessel positions are live AIS, importers come from real customs records (pre-arrival matches are labeled inferred), and costs come from a published tariff. Band is the part that makes it work: the room, the runtime recruitment, the @mention handoffs, and the audit trail all run on Band.

VERITAS

VERITAS

VERITAS is a multi-agent content verification system built to fight misinformation at scale. Every day, billions of unverified claims circulate online. Manual fact-checking takes hours — and most of the time, it never happens. VERITAS automates this process entirely, turning a hours-long task into under 60 seconds. The system accepts any block of text and returns a fully sourced, trust-scored verification report for every factual claim it contains. The pipeline runs through seven specialized AI agents coordinated exclusively through Band. The Orchestrator extracts every verifiable claim from the input and forwards them to ResearcherMaster. ResearcherMaster divides the claims into four groups and dispatches them simultaneously to ResearcherB, ResearcherC, and ResearcherD — triggering genuine parallel research across four independent agents, all powered by AIML API and searching the web in real time via Tavily. Source Validator then collects all findings, evaluates every source for credibility, and filters out blogs, forums, and unreliable sites. Finally, Reporter synthesizes the validated findings into a structured report: a trust score from 0 to 100, a plain-language explanation, and verified citations for every claim. Band is not a wrapper in this system — it is the coordination backbone. Every task assignment, research finding, validation result, and final report passes exclusively through Band's @mention routing layer, enabling true agent-to-agent collaboration across the full pipeline. VERITAS targets journalists, researchers, educators, and anyone who needs to quickly verify what they read online — making reliable fact-checking accessible, fast, and transparent.