model-evaluation

OpenLLM Monitor is a plug-and-play, real-time observability dashboard for monitoring and debugging LLM API calls across OpenAI, Ollama, OpenRouter, and more. Tracks tokens, latency, cost, retries, and lets you replay prompts — fully open-source and self-hostable.

Machine learning system for predicting genetic disorders using genomic, clinical, and demographic data. Implements robust preprocessing, feature selection, and multi-model classification (RF, XGBoost, LightGBM, CatBoost) with cross-validation to support early, data-driven genetic risk assessment.

A lightweight web viewer for LLM evaluation results — visualize, browse and analyze evalscope outputs instantly in your browser.

Business-specific AI benchmark packs for repeatable workflow regression checks.

ML-powered web app that scores academic references based on metadata (citations, year, authors, publisher) to support research quality.

IoT smart parking platform with ML-powered occupancy forecasting. React + Django REST API, scikit-learn Random Forest, dynamic pricing, and ESP32-ready sensor integration. Built for airport-style multi-zone parking (NMIA).

83 source-grounded, duplication-checked SVG components for LLM benchmarks, model launches, leaderboards, and paper figures.

Detects LLM prompt/model regressions before they ship — multi-provider eval, LLM-as-judge, drift monitoring, and CI/CD integration.

Dependabot for AI model dependencies: evidence-backed model upgrade PRs.

CatBoost-based housing price regression model with cross-validation and uncertainty estimation.

Compatibility profiles for entities joining human-AI collaborations

A curated collection of machine learning projects covering data preprocessing, model training, evaluation, and experimentation workflows.

LLM safety evaluation data workflow MVP with risk taxonomy, synthetic samples, quality gates, rubric judge, and Vercel demo.

Local bilingual LM Studio benchmark UI with matched German and English objective JSON suites.

Local-first CLI for real AI coding workflows: cost, retries, verification. No source or prompts uploaded.

🐍 A live visual arena where LLMs compete in Snake - compare model latency, reliability and decision-making across OpenAI-compatible APIs.

A streamlined and easy-to-use AI performance evaluation / summary template with modern UI in HTML, including correct percentage chart and comparison with other models, precision, recall, F1-score, and confusion matrix. Enables you to create the result chart within 3 minutes.

Benchmark, compare, and cost-predict your MCP agents — JMeter for AI agents.

🔍 Detect fraud in application data using machine learning and data visualization to uncover anomalies and enhance digital integrity.

Structured 10-step diagnostic for go/no-go feature evaluation in production ML pipelines. Covers outcome gradient, coverage gaps, entropy/gain ratio, conditional MI, incremental CV AUC, and 9-point temporal safety. Claude Code skill.