Gemma

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 date	February 2024
Author	Google DeepMind in collaboration with Google AI teams
Website	[Google AI Gemma]https://ai.google.dev/gemma
Repository	Google AI Developer Resources
Type	Open-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.

Edit on GitHub

Google Gemma AI technology Hackathon projects

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

TrapScan — AI Agent Trap Detector

TrapScan is a browser-native AI security tool designed to detect adversarial web content targeting AI agents in real time. As AI assistants increasingly browse websites, summarize information, and execute tasks autonomously, the web itself is becoming an attack surface. Attackers can embed hidden instructions inside webpages that are invisible to humans but fully readable by AI systems. These attacks can manipulate AI reasoning, inject malicious prompts, override safeguards, or trigger unauthorized actions. Inspired by Google DeepMind’s “AI Agent Traps” research, TrapScan implements detection for six major AI agent attack categories, including prompt injection, semantic manipulation, jailbreak attempts, hidden behavioral control patterns, and systemic adversarial traps. TrapScan combines fast local browser detection with AI-powered classification using Gemma 4 (gemma-4-26b-a4b-it). The Chrome extension scans HTML, CSS, metadata, JSON-LD schema, hidden DOM elements, and suspicious prompt patterns directly inside the browser. Suspicious findings are then analyzed by Gemma 4, which classifies threats, assigns risk scores, filters false positives, and explains attacks in plain English. The project includes: * A Manifest V3 Chrome extension * Real-time threat analysis UI * Browser risk indicators * Scan history and downloadable audit reports * A live Vercel-hosted web demo for instant testing TrapScan represents a new category of browser-native defense tooling focused specifically on protecting AI agents from manipulation on the open web.

AMS-Gemma4: Offline Antibiotic Stewardship AI

AMS-Gemma4 is an offline clinical decision support system for antibiotic stewardship in Indian hospitals. Antibiotic resistance kills more people than HIV and malaria combined. India is at the centre of this crisis. The system runs entirely on a standard laptop via Ollama — no internet, no cloud, no patient data leaving the device. Built on Gemma 4 E2B with native function calling architecture. Key features: - Python function calling: calculate_crcl(), check_aware_tier(), lookup_resistance() - Clinical safety gate: automatically rejects contraindicated drugs - WHO AWaRe 2023 classification built-in - ICMR India 2024 resistance patterns - Cockcroft-Gault renal dose adjustment - Tamil and English output - Multimodal: upload culture report photo Clinical validation: UTI case: Nitrofurantoin REJECTED at CrCl 31.5 mL/min. Ceftriaxone recommended correctly. MRSA case: Doxycycline preferred over Vancomycin per AWaRe stewardship principles. Built by Prakash Raj K, Associate Professor of Pharmacy, Sri Balaji Vidyapeeth, Puducherry, India.

Trellis: The Knowledge Fabric for Law Firms

Every law firm sits on an unmined fortune: the experiential wisdom of senior partners. Yet, when a veteran attorney retires or departs, decades of strategic instinct permanently evaporate. This institutional knowledge loss costs mid-to-large firms $15M to $40M annually. The barrier to documentation is economic. Lawyers bill in six-minute increments; spending time writing down lessons represents an immediate loss of billable revenue. While generative AI should solve this, strict obligations toward attorney-client privilege block public LLM tools. Meanwhile, vertical legal tools like Harvey provide generic legal advice but lack a firm's specific internal wisdom. Trellis solves this with a secure, two-tier system architecture. The first layer is the Personal Second Brain. This is a private, on-device edge environment running Gemini Nano where attorneys quickly capture unstructured thoughts via voice memos, notes, or image OCR. Because data stays local, capture is entirely unredacted and factual. The second layer is the Team-Managed Knowledge Graph. This shared ledger is built through an automated, dual-pass sanitization pipeline. First, Microsoft Presidio strips explicit PII like names, dates, and entities. Second, Gemini Pro abstracts the specific case details into generalized strategic principles. The lawyer reviews a side-by-side diff before approving publication to the firm's shared memory. Using a hybrid vector-graph RAG model, team members can query this collective intelligence using natural language. To eliminate legal risk, Trellis enforces a hard deterministic guardrail: if the retrieved context score falls below a strict threshold, the system executes a strict refusal rather than risk a hallucination. Finally, Trellis exposes a Model Context Protocol (MCP) endpoint. This allows third-party tools like Harvey, CoCounsel, or Copilot to securely plug directly into the substrate, grounding generic legal AI capabilities in the firm's actual historical wisdom.

Vibe Studio: AI-Native Agentic IDE

Vibe Studio represents a paradigm shift in how developers interact with Artificial Intelligence. Most current solutions are merely basic chat wrappers grafted onto existing code editors, lacking true integration with the development lifecycle. Vibe Studio, however, is built from the ground up as an AI-native Integrated Development Environment featuring a custom Agentic UI. At its core, Vibe Studio replaces the standard "prompt-and-pray" approach with a deterministic, event-driven architecture. We implemented an asynchronous Pub/Sub execution bus and a rigorous state machine (RoadmapTracker) that orchestrates complex, multi-step LLM workflows. This means the AI doesn't just spit out code snippets; it follows a structured lifecycle—from architectural analysis to implementation, and verification. Our platform is designed for senior engineers who demand control. The human architect sets the constraints, defines the patterns, and dictates the workflow, while the autonomous agents handle the mechanical execution. Features include real-time intent-guard pipelines to prevent prompt contamination and strict architectural enforcement. Vibe Studio isn't just another coding assistant; it is a robust, production-grade infrastructure designed to scale developer output without sacrificing engineering fundamentals.

AuditForge

Enterprise AI deployments are moving fast, but the systems those agents touch — codebases, APIs, databases — are still audited manually, slowly, and inconsistently. AuditForge changes that. AuditForge is a multi-agent compliance audit platform built on Google Gemini. Security teams upload their artifacts — a codebase, an OpenAPI spec, a database schema, a cloud config — and AuditForge dispatches specialized Gemini agents to analyze them against OWASP API Top 10, HIPAA Technical Safeguards, and SOC2 Common Criteria. Gemini's long-context window is the core advantage: rather than scanning files in isolation, the analysis agent reads an entire codebase at once, catching cross-file vulnerabilities that line-by-line tools miss entirely. Every finding is standardized — severity, evidence with the exact file and line, AI-generated remediation with a code example, and a direct mapping to the compliance clause it violates. When the audit is ready to hand off, AuditForge generates a cryptographically signed PDF report that maps every issue to its regulatory reference — the kind of document a CISO or external auditor can act on without a translator. Adding support for new system types — Terraform, Kubernetes, CI/CD pipelines — requires only a new connector module; the policy engine and report layer need no changes. Every action is recorded in a tamper-evident, cryptographically chained audit trail, making the audit of the audit verifiable too. The frontend runs natively on desktop and in the browser from a single Kotlin Multiplatform codebase. Server-Sent Events stream findings in real time as the scan runs. AuditForge makes compliance auditing something engineers can run themselves, security teams can trust, and regulators can read.