Chroma

Chroma is building the database that learns. It is an open-source AI-native embedding database. Chroma makes it easy to build LLM apps by making knowledge, facts, and skills pluggable for LLMs. The fastest way to build Python or JavaScript LLM apps with memory

General
Relese date	2023
Author	Chroma
Type	embedding database

Tutorials

Great tutorials on how to build with Chroma

👉 Discover more Chroma Tutorials on lablab.build

Chroma - Helpful Resources

Check it out to become Chroma Master!

Chroma Chroma website
Chroma Docs Documentation of Chroma
Chroma Repo Open-source Chroma repository

Chroma - Clients

Connect Chroma to your Project!

Chroma Python Python Client for Chroma
Chroma JavaScript High-Performance Chroma Client for JavaScript

Chroma - integration

Get started with Chroma!

DocuPilot AI: Smart Docs & AI Asst.

DocuPilot AI by Tech_Innovators streamlines document processing and knowledge management for businesses. It tackles manual data extraction, inefficient searching, and verification challenges in documents like invoices, contracts, and reports. The platform integrates document extraction, intelligent search, invoice analysis, automatic classification, and human review. Users can upload documents, extract text, query content, and inspect original evidence securely. Key Features: 1. **Intelligent Document Extraction:** Extracts text from PDFs, images, and TXT. 2. **Evidence-Grounded Q&A:** Provides answers with source references and evidence. 3. **Invoice Intelligence:** Extracts invoice data, performs checks, and generates financial reports. 4. **Explainable Document Classification:** Auto-categorizes documents with traceable evidence. 5. **Human-in-the-Loop Verification:** Allows users to correct and approve extracted data. 6. **Private Document Management:** Ensures security with authentication and authorization. 7. **Optional Asynchronous Processing:** Supports background extraction with job tracking. Built with modern web technologies, DocuPilot AI emphasizes trustworthy workflows by linking information to original evidence and enabling human verification. It's ready for hackathon submission and demonstrations; public deployment requires further security validation.

MediGuide AI Multilingual Healthcare Intelligence

## MediGuide AI – Multilingual Healthcare Intelligence MediGuide AI is an AI-powered healthcare accessibility platform designed to help patients understand complex medical reports in their preferred language. ### The Problem Medical reports often contain complex terminology that patients struggle to understand. Language barriers can make healthcare information even more difficult to access, affecting patients' ability to prepare for consultations and ask informed questions. ### Our Solution MediGuide AI transforms medical documents into more accessible, patient-friendly information through an intelligent document-processing pipeline. ### Key Features - **Medical Document Extraction:** Uses PaddleOCR and text extraction to process uploaded medical reports. - **Clinical Entity Recognition:** Identifies important medical terms and entities using a pretrained biomedical NER model. - **Report Summarization:** Produces concise summaries to make medical information easier to review. - **Multilingual Accessibility:** Supports English, Tamil, Hindi, Telugu, Malayalam, and Kannada. - **AI Health Assistant:** Helps patients understand medical terminology and prepare questions for their doctors. - **Emergency Awareness:** Detects potentially urgent symptom descriptions and advises users to seek appropriate medical care. ### Technology Stack - Frontend: React - Backend: Python, FastAPI - OCR: PaddleOCR and PyMuPDF - NLP: Transformers and biomedical named entity recognition - Database: SQLite - Multilingual processing: Local language detection and translation components ### Innovation and Impact MediGuide AI combines medical document processing, clinical NLP, summarization, and multilingual accessibility in one platform. It aims to reduce language barriers and help patients participate more confidently in healthcare discussions. MediGuide AI is an educational support tool, not a replacement for professional medical advice, diagnosis, or treatment.

TradeForensics: Commerce Action Copilot

TradeForensics is an evidence-grounded operational diagnostics engine and action copilot built for high-volume e-commerce marketplaces, evaluated across 98,666 real-world orders. Modern commerce operators are paralyzed by disconnected signals: financial transactions in SQL tables, carrier delivery milestones in logistics systems, and unstructured reviews in customer service inboxes. Standard AI tools fail because vector embeddings cannot calculate statistics, leading to hallucinated numbers and un-auditable advice. TradeForensics solves this through a 4-tier hybrid forensic architecture: 1. Deterministic Analytics Engine: Calculates baseline KPIs and discovered the 5.75x dissatisfaction multiplier (late shipments cause low-rating rates to jump from 9.21% to 52.98%). 2. NLP Portuguese Topic Classifier: Categorizes complaints across 4 core operational failure modes at 97.00% accuracy. 3. LightGBM Tabular Risk Attribution: Predicts low-score dissatisfaction risk with 0.9071 ROC-AUC, isolating delay days and freight ratio as primary drivers. 4. ChromaDB Qualitative RAG: Retrieves authentic customer complaints to provide verifiable proof. A bounded Action Copilot synthesizes prioritized operational remediation checklists with Low/Medium/High effort ratings without mathematical hallucinations. Engineered for AMD Instinct MI300X accelerators with 192GB HBM3 memory and 5.3 TB/s bandwidth, TradeForensics achieves 208ms Time-To-First-Token and 68 tokens/sec throughput, enabling high-density multi-seller batch forensics.

AMD Hardware Advisor

AMD Hardware Advisor is an AI-powered hardware research assistant designed to help developers and AI/ML engineers make informed decisions about AMD hardware. Using Retrieval-Augmented Generation (RAG), the application retrieves relevant passages from official AMD documentation to answer technical questions with traceable source citations. Users can explore a structured hardware catalogue, compare supported products side by side, and receive workload-based hardware recommendations with clear rationales. The project combines a Python backend, local semantic embeddings, a persistent vector database, and configurable LLM inference. Its key focus is trustworthy, source-grounded answers rather than unsupported hardware specifications. The system also includes an AMD inference-provider integration and verification mechanism to distinguish confirmed AMD execution from an unverified or local fallback. AMD Hardware Advisor brings hardware research, technical Q&A, product comparison, and AI workload selection into one unified interface.

GENVEY=gamified surveys

Traditional customer and market research surveys suffer from abysmal 2% completion rates, rampant cognitive fatigue, and low-conviction answers. Our platform replaces static forms with a Neuro-Symbolic AI Game Compiler that converts natural-language business intent and demographic targets into deep, mechanically rich 3D and 2D games. Powered by Gemma-4-31B-IT running on AMD Token Factory cloud GPUs, the compiler operates with zero local VRAM and sub-2-second generation latency. When a studio or enterprise specifies a target audience (e.g. "Gen Z competitive souls gamers") and survey intent (e.g. "Assess pricing willingness for a $10 sovereign tier vs $2 starter tier"), our persistent ChromaDB Trend RAG matches real-time gaming sentiment while our Cohort Playtime Profiler selects the winning gameplay format based on empirical session durations. The engine compiles a cohesive 4-zone interactive game world: 1. Courtyard & Forge Hub: Dual physical survey forges ($10 Sovereign Anvil vs $2 Adventurer Anvil) where players verify and spend earned gold, directly quantifying purchase intent through gameplay. 2. Catacombs & Lever Gate: Mechanical levers that physically lift iron portcullises, solid AABB Box3 sliding collisions (zero walk-through walls), treasure chests granting gold rewards, and destructible explosive barrels. 3. Faction Sanctum: Shrines and D20 skill check dialogues with Envoy NPCs. 4. Cathedral Arena: Multi-phase boss battle with Active Parry (200ms timing window), Critical Ripostes, stamina management, and target lock-on. Players can experience the game immediately in 100% true fullscreen browser WebGL via Three.js or launch it as a native desktop application in Godot 4.7 Engine with native GDScript combat state machines. Brands achieve 85%+ completion rates and unvarnished System-1 behavioral telemetry, turning survey fatigue into high-engagement interactive play.

RAG-MATCH3-AMD

RAG-Match3-AMD is an intelligent document question-answering system designed to help users find reliable information inside PDF, DOCX, spreadsheet, and text documents. The project uses Chroma Cloud as its vector database and retrieval engine. Documents are parsed, divided into manageable chunks, and transformed into dense embeddings with Chroma Cloud Qwen and sparse embeddings with Chroma Cloud SPLADE. Hybrid search and Reciprocal Rank Fusion combine semantic and keyword results to retrieve the most relevant evidence. OpenAI GPT-4.1 Mini then analyzes the retrieved context and generates a concise answer supported by citations to the original document. The system can also process images and scanned content through a vision-capable model. Document metadata, chunk indexes, and source identifiers are preserved to improve traceability and prevent duplicate results. The project provides a practical RAG workflow that separates document retrieval from answer generation. Chroma Cloud handles embedding and search, while OpenAI handles language understanding and response generation. This architecture makes the application suitable for learning materials, technical documentation, reports, and other knowledge-intensive document collections