BreezeBlue's Breeze TTS 2 Clones and Designs Voices in Under 40ms
BreezeBlue open-sourced a 3B bilingual TTS model that tops the Artificial Analysis leaderboard with sub-40ms latency and natural-language voice design.
// Intelligence
Open Source
A focused feed of models, agents, research and open-source releases for people building with AI.
Updated
Zoetrope Turns Chaotic Claude Code Sessions Into Live Flow Graphs
A new open-source Rust tool renders Claude Code sessions as a live, scrubbable flow graph inside your terminal or browser.
Apodex's FrontierAgent Lets a 35B Model Match 1T-Parameter Rivals
Apodex open-sourced FrontierAgent, an Apache 2.0 terminal agent framework with ReAct and Agent Team modes, alongside a 35B open-weight model.
Kyutai Opens Pocket TTS Training So Anyone Can Build CPU Voice Cloning
Kyutai released the full training stack for its 100M-parameter Pocket TTS, letting anyone train a voice model from scratch for under $200.
Z.ai's GLM-5.3 Got So Dangerous It Delayed Its Own Release
Z.ai's new 753B open-weights model keeps the GLM-5.2 base but doubles down on post-training, topping open-source coding and cyber benchmarks.
STUDIO 1939 Turns MiniMax H3 Into a 1930s Animation Studio
A style LoRA for MiniMax H3 generates video with hand-painted gouache backgrounds and golden-age character animation from text alone.
Mira Murati's Thinking Machines Pays Researchers $50K to Make Open Weights Safer
Thinking Machines Lab is offering up to $50,000 in Tinker credits to researchers tackling the hardest open-weight model safety problems, from tamper-resistant safeguards to worst-case risk forecasting.
SparkLLM's Spark-X2.5 Packs 1M-Token Context Into a Tiny 4B Model
A compact 4B open-source model with hybrid sliding-window attention, native 1M-token context, and agent-focused benchmarks that top comparable small models.
Alibaba's MiniMax-H3-Fun-Controlnet-Union Steers Video Generation With One 6.8 GB Checkpoint
Alibaba PAI drops a single ControlNet-Union checkpoint that adds Canny, Depth, HED, MLSD, Pose and inpainting to MiniMax-H3 video generation.
Someone Reverse-Engineered Grok Bot 0.18 and Rebuilt It With Four AI Backends
A weekend reverse-engineering project turns a compiled macOS app into 440K lines of readable TypeScript, then bolts on a four-provider inference router.
Codex Router Merges DeepSeek and Kimi Into OpenAI's Picker at 89x Lower Cost
A local router merges Kimi, DeepSeek, and other external models into the Codex picker without kicking out your native GPT models or ChatGPT login.
IBM's Docling Hits 65K Stars Fixing the PDF Problem That Breaks RAG
The open-source document parser Docling has crossed 65,000 GitHub stars, cementing its place as a go-to tool for prepping files for gen AI pipelines.
Inbox Zero's Rakazo Lets You Self-Host AI Agents Without Vendor Lock-In
Rakazo bundles persistent AI teammates with their own live Linux desktops into a self-hosted stack you run on your own hardware and model keys.
Code-Graph-RAG Gives AI Agents a Real Map to Stop Hallucinating Monorepos
An open-source tool parses your monorepo with Tree-sitter, builds a Memgraph knowledge graph, and hooks into Claude Code via MCP.
TrueFoundry's TrueForge Open-Sources an Agent Runtime That Cuts Claude Costs by 75%
TrueFoundry open-sourced TrueForge, an MIT-licensed agent runtime that ran 30-75% cheaper than Claude Managed Agents on a 14-task enterprise benchmark.
DeepSeek Harness Hits 160K Stars Challenging Claude Code's Closed Stack
DeepSeek open-sourced dsh, a plugin-based coding agent framework where the model, tools, sandbox, UI, and even the agent loop itself are all swappable.
Block Ships Berd, an Open-Source Desktop Hub for Managing AI Agents
Block open-sourced Berd, a Tauri-based desktop shell that unifies AI agents, projects, and files across any model provider under Apache 2.0.
Colibri Runs a 744B Model on a Laptop With No GPU
A single-file C engine streams experts from disk to run 744B-parameter MoE models on a 25GB laptop, no GPU required.
Hugging Face's Tau Teaches Developers How Coding Agents Actually Work
Hugging Face released Tau, an MIT-licensed terminal coding agent designed as much for reading as running, with a three-layer architecture you can actually understand.
Stanford's Marin Trains a 535B Open Model Anyone Can Watch Live
Stanford's Marin lab kicked off a 535B mixture-of-experts run with 23B active parameters, streaming every metric, config, and mistake live to the public.
Kizuna AI Lab's Sokuji Translates Live Calls Offline Without a GPU
Sokuji is an open-source live speech translator that runs entirely on-device with WASM and WebGPU, no API key or GPU required.
Magnitude Turns Local AI Model Selection Into a One-Click Decision
Magnitude's new catalog profiles your hardware, estimates tokens per second for every model, and picks the best local LLM before you download anything.
Qwen3.8-27B Unleashed Hits 83% MMLU Uncensored on a Single 24GB GPU
A per-tensor dynamic quantization of an abliterated Qwen3.8-27B lands on Hugging Face, delivering 82.98% MMLU and 262k context on a single RTX 4090.
Kyutai's MuScriptor Now Turns Any Song Into Printable Sheet Music
Kyutai's open-source music transcription model can now export sheet music PDFs, editable MusicXML, and guitar tabs alongside MIDI output.
No stories match these filters.