AI Models

AI Models OpenAI’s New Open-Weight Models: gpt-oss-120b & 20b

OpenAI introduces gpt-oss-120b and gpt-oss-20b, two open-weight models released under the Apache 2.0 license. These models offer advanced reasoning, native tool integration, and flexible deployment options ranging from production clusters to local hardware.

AI Models ShareGPT-4o-Image & Janus-4o: Open-Source Models Reaching GPT-4o Output Quality

ShareGPT-4o-Image provides 92k high-quality samples generated by GPT-4o. Janus-4o, fine-tuned from Janus-Pro, introduces text-plus-image to image generation capabilities. Includes complete code for inference and training.

AI Models TensorZero: Optimize LLM Applications with Production Feedback

TensorZero integrates an LLM gateway, observability, and optimization tools. Leverage production metrics and human feedback to refine prompts, models, and inference strategies.

AI Models Xiaomi MiMo-7B: Built From Scratch for Math and Code Reasoning

Xiaomi’s MiMo-7B series outperforms larger models on math and coding benchmarks. Trained from scratch using MTP and RL, it rivals OpenAI o1-mini. Available now on HuggingFace.

AI Tools Earth Copilot: Query Geospatial Data Using Natural Language

Earth Copilot converts natural language queries into map visualizations and actionable insights. Query satellite data, analyze climate trends, and search private catalogs. Includes a full setup guide.

Developer Tools Fuck-U-Code: A Brutally Honest Code Quality Analyzer

Use this local CLI tool to identify the messiest parts of your codebase across Go, Python, Java, and more. It scores your technical debt on a 0–100 scale and generates colorful, unfiltered reports.

Music Players Infinite Radio: The AI DJ That Adapts Music Genres to Your Screen

Infinite Radio combines Magenta’s real-time music model with active screen analysis. Dynamically change genres based on your open applications or visual display content. Runs locally via Docker.

Text Tools Paperless GPT: Smarter OCR and Auto-Tagging for Paperless-NGX

Enhance Paperless-NGX with AI-driven OCR and automated tagging. Fix illegible scans using GPT-4o or Ollama and eliminate the need for manual filing.

Voice & Speech Tools ChatTTS: A Text-to-Speech Model Optimized for Dialogue

ChatTTS generates natural, expressive speech with precise control over laughter, pauses, and tone. Explore code examples, hardware requirements, and speaker embedding techniques.