Run Stable Audio 3 Locally on a Raspberry Pi With aria
aria is a dependency-free native runtime that runs the full Stable Audio 3 text-to-music pipeline on ordinary GPUs, CPU-only laptops, and an 8GB Raspberry Pi 5, no Python required.
Run Any Hugging Face Model at vLLM Speed
Hugging Face and vLLM shipped a native-speed transformers backend on July 8, 2026, serving almost any Hugging Face model at full vLLM speed with a single flag.
Otari: Self-Hosted OpenAI-Compatible LLM Gateway
Mozilla AI has released Otari, an open-source, self-hosted LLM gateway that puts a single OpenAI-compatible endpoint in front of more than 40 model providers.
Rowboat: Open-Source Local AI Coworker vs Claude Cowork
Rowboat is a free, open-source, local-first AI coworker built as a Claude Cowork alternative, running any model via Ollama or LM Studio.
Cohere Transcribe Arabic: Open ASR Beats Whisper
Cohere released Transcribe Arabic, an open-weight Apache 2.0 speech-to-text model built for Arabic dialects, bilingual speech, and code-switching.
NVIDIA Audex: Open Audio LLM for Speech and Sound
NVIDIA has released Audex, an open-weight audio-text model that does speech recognition, translation, text-to-speech, and general audio generation in one network.
PocketTTS-RAVEN: In-Browser TTS With Voice Cloning
PocketTTS-RAVEN runs text-to-speech and voice cloning entirely in a browser tab, with no server, GPU, or account, at faster-than-realtime speed.
Tencent Hy3: Open 295B Agentic Coding Model
Tencent open-sourced Hy3, a 295B mixture-of-experts model (21B active, 256K context) under Apache 2.0. It beats GLM-5.2 at roughly half the size and is free on OpenRouter until July 21.
Ternlight: 7MB Embedding Model Runs in the Browser
Ternlight packs semantic search into a 5 to 7 MB WebAssembly bundle that runs on the CPU with no API, and it hit the Hacker News front page within a day of launch.
LongCat-2.0: China's 1.6T Open-Weights Coding Model
Meituan open-sourced LongCat-2.0, a 1.6-trillion-parameter agentic coding model trained entirely on Chinese chips. The weights just landed on Hugging Face, and it beats GPT-5.5 on SWE-bench Pro.
NVIDIA Nemotron TwoTower: 2.4x Faster Open Diffusion LLM
NVIDIA's Nemotron-Labs-TwoTower generates text 2.42x faster at 98.7% quality by adding a denoiser tower to a frozen autoregressive backbone, no retraining required.
First Open-Source Diffusion ASR Model: How It Works
Interfaze's diffusion-gemma-asr-small is billed as the first open-source diffusion-based speech recognition model, refining random tokens into a transcript instead of decoding left to right.
ZeroLabs: Free Open-Source ElevenLabs Alternative
ZeroLabs stitches six open-source models into one free browser studio for voice cloning, design, cleanup, sound effects, and transcription, running on Hugging Face ZeroGPU.
Claude-Real-Video: Let Any LLM Actually Watch a Video
claude-real-video is a local CLI that pulls scene-change frames, dedupes repeats, and transcribes audio so Claude, ChatGPT, or Gemini can actually watch a video, not just read its transcript.
ZCode: Zhipu's GLM-5.2 Coding Agent, Chat Triggers
Zhipu has shipped ZCode, an official desktop coding agent built on its open-weight GLM-5.2 model, and it lands with a feature no other coding agent ships today: you can trigger and advance tasks from a chat app.
ClinePass: Open-Weight Coding Models for $9.99 a Month
Cline, the open-source AI coding agent trusted by 8M+ developers, launched ClinePass, a flat $9.99/month subscription bundling curated open-weight coding models directly into its IDE, CLI, and SDK.
ComfyUI-Angelo Adds Outpainting and 2x Upscaling
ComfyUI-Angelo v2.2.1 adds one-click outpainting and 2x super-resolution to its click-to-refine sampler for FLUX 2 Klein, Qwen-Image-Edit, and SDXL.
Kage gives AI agents verified, code-grounded memory
Kage, a new open-source verification layer built on Google's Open Knowledge Format, stops AI coding agents from recalling hallucinated or out-of-date memories.
Ploof: Agent-Native CLI for AI Image, Video, Audio
Ploof is a new open-source command-line tool that lets AI coding agents like Claude Code generate images, video, and audio without leaving the terminal.
DeepSeek DSpark Speeds Up V4 Inference by 80 Percent
DeepSeek shipped DSpark and open-sourced DeepSpec, a speculative decoding stack that speeds up V4 inference by up to 80 percent with identical output.
audio.cpp Runs 12 Audio AI Models in One C++ Engine
audio.cpp packs more than a dozen audio AI models into one ggml-powered C++ runtime, running TTS, voice cloning, and music up to 5x faster than Python.
Liquid AI LFM2.5-230M: On-Device Agent Model
Liquid AI's LFM2.5-230M is a 230M open-weight model small enough to run agentic tool-use directly on a phone or a Raspberry Pi.
Ornith 1.0: Open Agentic Coding Model on One GPU
DeepReinforce has released Ornith 1.0, an open-weights agentic coding model that runs on a single GPU yet posts frontier-level scores on agent benchmarks.
OpenKnowledge: Open-Source AI Wiki for Claude and MCP
OpenKnowledge is an open-source, local-first markdown editor and LLM wiki where Claude, Codex, and Cursor edit your notes directly through MCP.