Tag: Open Source AI

Discover AIRI, the self-hosted AI companion platform inspired by Neuro-sama. Build your own digital being with real-time voice chat, game-playing agents, and complete privacy. Runs in browser, desktop, and mobile.

Discover how AgentGym-RL trains 7B parameter LLM agents to surpass GPT-4o on real-world multi-turn tasks using progressive horizon scaling reinforcement learning. Complete setup guide, code examples, and performance benchmarks included.

DataFlex is an open-source dynamic training framework that revolutionizes LLM training through intelligent data selection, mixture, and reweighting. Built on LLaMA-Factory, it delivers proven performance gains while slashing compute costs.

VoxCPM2 is a tokenizer-free multilingual TTS system with 48kHz output, voice design from descriptions, and controllable cloning. Built on a 2B parameter diffusion autoregressive architecture, it supports 30 languages with production-ready deployment options.

Discover NVIDIA PersonaPlex, a real-time full-duplex speech-to-speech model with independent voice and role control. Learn installation, code examples, and why it's disrupting voice AI.

Heretic is a fully automatic censorship removal tool for language models that uses advanced directional ablation with TPE optimization. Run one command to decensor AI models while preserving their intelligence. Over 3,000 community models created.

dot-skill transforms colleague communication data into interactive AI personas that preserve institutional knowledge. This open-source framework distills real cognitive patterns from chat logs, emails, and documents into installable Skills for Claude Code, Hermes, OpenClaw, and Codex.

Discover Parlor, the free on-device voice and vision AI powered by Gemma 4 E2B and Kokoro TTS. Run real-time multimodal conversations locally with zero API costs and complete privacy. Complete setup guide included.

Lucebox megakernel delivers 2× faster LLM inference on consumer GPUs through revolutionary single-kernel CUDA fusion. This deep dive covers installation, real code examples, benchmarks, and why developers are switching from general-purpose frameworks.

Discover Chatterbox TTS by Resemble AI—the open-source text-to-speech model with native paralinguistic tags like [laugh] and [cough]. Learn how its 350M parameter Turbo variant generates human-like speech in a single step, slashing compute costs while delivering production-quality audio for voice agents, gaming, and accessibility tools.

Discover PageIndex, the open-source vectorless RAG system achieving 98.7% FinanceBench accuracy without vector databases or chunking. Learn installation, code examples, and why reasoning-based retrieval outperforms similarity search.

Moonshine AI is the open-source voice toolkit delivering 100x faster on-device speech recognition than Whisper. With sub-100ms latency, cross-platform deployment, and built-in intent recognition, it's redefining what's possible for real-time voice interfaces.

Discover how FinGPT democratizes financial AI with open-source LLMs that outperform BloombergGPT at 1/10000th the cost. Complete guide with real code examples, benchmarks, and deployment strategies for developers.

HunyuanWorld-1.0 is Tencent's open-source framework for generating immersive, explorable 3D worlds from text or images. Learn installation, code examples, and why developers are switching from video-based tools.

Uzu is a high-performance inference engine that lets developers run AI models directly on Apple Silicon with zero latency, complete data privacy, and zero API costs. Built in Rust with bindings for Python, Swift, and TypeScript.

SimpleMem achieves 43% F1 scores on LoCoMo with 30× fewer tokens than full-context methods. This open-source memory framework for LLM agents supports text, image, audio, and video with seamless MCP integration for Claude, Cursor, and more.

Discover Everywhere, the open-source context-aware AI assistant that sees your screen in real-time. No screenshots, no app switching—just press a shortcut and get intelligent help integrated with multiple LLMs and MCP tools.

Discover Chandra OCR 2, the open-source model topping benchmarks for complex document layouts, handwriting, and 90+ languages. Complete installation guide, real code examples, and performance comparison inside.