nvidia-nemotron

The awesome collection of Nvidia NemoClaw presets, recipes, and playbooks for sandboxed OpenClaw operations.

AI-powered flood intelligence system featuring h2oGPTe Agent + NVIDIA NIM integration (A2A), NAT pipeline with Nemotron 49B, and real-time flood risk assessment. Part of NVIDIA–H2O.ai AI for Good Blueprint for disaster response and monitoring.

AI-powered business analytics agent built on 6 NVIDIA models: Ultra 253B, Nano 8B, Vision 12B VL, NV-EmbedQA-E5-v5 + self-hosted RTX 4090. 4-tier intent routing, PhD-level statistical analysis. Live at lubot.ai

NVIDIA Nemotron-Personas-Korea 합성 데이터셋의 통계 충실도·구조 독립 감사. Phase 1-3 (단변량/이변량/PGM 구조 추론 via CMI+PC) + 외부 모델 리뷰 키트 포함.

Recoup AI — self-improving, FDCPA-compliant AI voice agents for debt collection with a Salesforce-style CRM. Pipecat + NVIDIA Nemotron + Cekura (YC Voice Agents Hackathon).

NVIDIA Nemotron 11B Voice AI: Zero-Latency Local ElevenLabs Alternative - Real-time full-duplex speech AI pipeline with native tool calling.

Sample code for two-tier LLM inference (cascading) on Amazon Bedrock, using a single Bedrock Converse API surface. NVIDIA Nemotron Nano handles routine support-ticket classification on every request; Anthropic Claude Sonnet handles cases the routing logic flags as harder. Includes a bake-off harness, sample data, and tests. Next.js + TypeScript.

A production-ready Multi-Agent RAG system with 4 specialized agents (Retriever, Planner, Reasoner, Answer), Redis cache, SSE streaming, conversation memory, and comprehensive evaluation. Supports Persian/English queries with local LLM (Gemma 2) and free API (Nemotron 3 Ultra). Built with FastAPI, LangChain, ChromaDB, and BGE-M3 embeddings.

Training and dataset-building pipeline for NVIDIA Nemotron reasoning experiments using Unsloth, TRL and NeMo RL.

Agentic backend for coordinating Toronto shelter capacity — pgvector tool routing, a 31-tool FastMCP server, and an A2A mesh of Nemotron-3 Nano agents. NVIDIA Spark Hackathon 2026.

Fully local CPU-only voice assistant using NVIDIA Nemotron ASR, Ollama/Gemma, Piper TTS, FastAPI and ALSA.

arXiv research briefing pipeline that routes natural language queries to relevant papers, caches PDFs in Backblaze B2, extracts key sections with PyMuPDF, and generates citation-backed insight briefs using NVIDIA Nemotron.

Token-budgeted 4-stage audit pipeline: AI checklist generation, observation optimizer, severity scoring, and executive summary, each enforcing a real tested 8k-token budget. FastAPI + React + NVIDIA Nemotron.

큐레이터를 위한 한국인 페르소나 생성기 (NVIDIA Nemotron-Personas-Korea 기반)

MindMap is a digital journaling platform for mental wellness support using Natural Language Processing and Generative AI.

Lyra Compliance AI is an AI powered compliance copilot that helps Swedish SMEs understand VAT, corporate, employment, and regulatory requirements through a conversational interface

Enterprise agent that fights wrongful insurance denials — grounds every argument in the insurer's own policy. Winner, RAISE Summit Hackathon 2026 (Vultr track).

Compare how vision models reason about images — not just their accuracy scores

Adaptive Token Optimization System — Middleware for reducing prompt token usage and inference cost through complexity-aware prompt optimization and model routing.

Checks every line on a medical bill against real Medicare rates and CMS coding rules, flags the charges that look wrong, and drafts the dispute letter. Built with LangGraph, NVIDIA Nemotron, FastAPI and Next.js.