Code for "Counterfactual Token Generation in Large Language Models", Arxiv 2024.

DelphiMistralAI wrapper brings Mistral’s text-vision-audio models and agentic Conversations to Delphi, with chat, embeddings, Codestral codegen, fine-tuning, batching, moderation, async/await helpers and live request monitoring.

A Go port of bgreenwell/doxx - A lightning-fast, terminal-native document viewer for Microsoft Word files.

A minimal Mistral API wrapper to OCR images and PDFs with images embedded in the markdown result.

A cloud native platform for service engineers and technicians to efficiently manage, track & resolve appliance service requests with real-time insights

ToneSlider is a sleek tone adjustment tool that lets users rewrite text with varying levels of formality using a simple slider interface. Powered by Mistral AI and optimized with Upstash Redis caching, it features smooth undo/redo functionality and persistent state management for a seamless user experience.

"Blog with Mistral GenAI" Project with Node.js for the backend and React for the frontend - This web application is a blog with a generative AI. To use it, you need to enter your Mistral API key in the .env file in the backend.

📸 Capture and share your emotions with a simple click. Let bh.AI analyze your expressions along with the chat and respond with personalized support. 🤖💬

Narrate AI is a demo application designed to generate stories, blogs, or journal entries based on key points or summaries, showcasing the power of Mistral AI for content creation.

A web application that outputs weird-yet-meaningful "never-before-thought" ideas. Built with FastAPI backend, Jinja2 Templates (Bootstrap 5 & NES.css) for UI, Gemini (1.5, 2.0, and 2.5 Flash Preview) for assumption mining, idea composition, and final polish, and Mistral-small for coherence & novelty scoring.