Tools Bench.

pytorch/pytorch

☆104.4K

Tensors and Dynamic neural networks in Python with strong GPU acceleration

morluto/rea

☆95.9K

Reverse engineer anything with agents, from app behavior down to native binaries.

bilawalsidhu/gods-eye-view

☆51.2K

A spy satellite simulator in your browser, except the data is real. Live open source spatial intelligence on a photorealistic 3D globe.

cathrynlavery/diagram-design

☆50.0K

Editorial diagram design for Claude Code, Codex, GitHub Copilot, Cursor, Factory Droid, and Pi. 44 diagram types. Self-contained HTML + SVG. No shadows. No Mermaid slop.

lyogavin/airllm

☆36.3K

AirLLM 70B inference with single 4GB GPU

storytold/artcraft

☆16.9K

ArtCraft is an intentional crafting engine for artists, designers, and filmmakers

ReSO AI

462

ReSO is a generative engine optimization platform that helps brands show up in AI search results. It’s built for founders, CMOs, and growth teams who want to turn buyer intent in ChatGPT, Perplexity, and Google AIO into visibility, citations, and traffic. What makes it different is that it doesn’t just track rankings, it analyzes intent, competitors, content gaps, and technical issues, then turns them into priority actions to improve AI recommendations.

Onepin

442

AI voices sound human now. But they still guess names from the spelling, and nobody has time to listen to every line. Onepin runs on the voices you already use (ElevenLabs, OpenAI, Google, 30+ more), spells out prices and dates, checks names against a 4M-word pronunciation dictionary, scores every line, and fixes the wrong word without re-rendering the take. Built for teams shipping AI voiceover in several languages: product videos, dubbing, courses. Free to start, no credit card required.

Microsoft-Decision-1

174

Microsoft-Decision-1 is a decision-scoring model for routing, classification, verification and agent control. Give it fixed options and it returns a calibrated probability for each in a single pass, as structured output software can act on directly. Post-trained from Qwen3.5-9B, it is about 35x faster than GPT-6 Sol at median latency, per Microsoft. Available now in Microsoft Foundry at $0.042 per million input tokens, with free output.

Skreno

93

Skreno is a video studio in your browser: record, edit and share in one place. Record your screen and camera as separate tracks, then edit in Studio: pick a layout, cut silences, and add animated captions and titles. Share one link with a transcript, AI summary and comments on the exact second. Send a request link and anyone can record for you, no account needed. Bug reports come with the console log attached.

Maildun for Mac

91

Maildun for Mac is an agentic email builder for macOS. Design by hand in the drag-and-drop editor, or connect Claude, ChatGPT, Gemini or Grok with your own API key: the assistant edits your design live, as a proposal you accept or reject. Turn on the built-in MCP server and Claude Code, Codex or Cursor can create and edit emails too. Plus workspaces per client, brand kits with shadcn-style component styles, saved sections, media on your own R2 or S3 bucket, and publishing to maildun.com.

Cue

158

Your headphones become a focus switch. Put your AirPods in and your Mac turns on your Focus mode, blocks distracting sites and starts a timer in the notch. Take them out: break time, sites come back. Also in the notch: now playing, volume and brightness, your own colors, session history and two tiny buddies that react to everything. Free, private, macOS 14+ on Apple Silicon. Vibe-coded by one person and Claude.

Maritime: Fully managed cloud for running AI agents at scale

19

Run millions of agents, each on its own isolated, persistent VM. Maritime · Fall 2026 · B2B Tags: Developer Tools, Cloud Computing, Infrastructure. Website: https://maritime.sh/

NanoMuse – An open-source AI agent for your phone and computer

65 pts

Hacker News discussion (65 points, 21 comments).

Durable Actors – OSS Durable Objects with configurable compute

52 pts

Hi HN, we're Thomas and Olivier from Terse ( https://www.useterse.ai/ ) We've built Durable Actors, an open-source alternative to Cloudflare's Durable Objects. A Durable Object/Actor is a tiny server that handles one request at a time and has its own SQLite database. There's exactly one of each in the world and it is addressed by name. This is the perfect primitive for deploying multiplayer agents. Each agent can have its own Durable Actor, and each user can connect to that Actor via websocket.... (52 points, 25 comments).

Breadcrumb, record everything on your mac + context manager for AI

49 pts

Hi HN, I'm Justin. Breadcrumb records everything you do on your Mac (screen + meetings + AI transcripts + what you and your AI decided) and turns it into memory your AI can search. It's local and encrypted. You can also teach it rules by talking to it and it makes sure the right rules turn up in the right context. Works with Claude Code / Codex / Cursor / opencode. All of this is exposed to your AI as 30+ MCP tools (here's the definitions): https://innerloop.works/breadcrumb/mcp I started it in... (49 points, 9 comments).

ManimGX – Manim's API on a Rust/wgpu engine, for 3D videos in Python

8 pts

Video as code took off with the Opus 5.5 release. People now make videos by writing code, with Remotion, HyperFrames and others. We wanted the same for 3D videos. You can use something like Blender, but it's heavy to install and slow to render. So we made ManimGX: - Lightweight: a 65–110 MB download, depending on the OS - Easy to install: pip install manimgx (Python 3.13+), with no system dependencies - Cross-platform: macOS, Windows and Linux, and even the browser via Pyodide and WebGPU - Blazi... (8 points, 1 comments).

Proton Drive for Linux

118 pts

Hello everyone, I wanted to share this small project that I have. It's born out of a necessity that I had, because Proton doesn't ship a Linux version of the drive yet (it's apparently coming out later this year but who knows?) and all the current solutions are hacky and wonky at best. But for the time being I really needed to mount my Proton drive on my Linux machine the way I mounted it on my Mac, so I (and my faithful AI slave) did this small utility that allows you to mount your Proton Drive... (118 points, 43 comments).

World Watcher – a map to seamlessly watch public webcam streams

8 pts

World Watcher is an interactive map of live webcams around the world, that prioritizes smooth UX and correct geolocation of cameras. The idea is to have a better experience of navigating live webcam streams that are publicly available on YouTube. There were already a few livecam aggregators out there that include a map, but I found none of them satisfying to use, as they always require you to open new pages to watch the streams. On World Watcher, you can jump from place to place seamlessly. New... (8 points, 2 comments).

MiniMax H3 Turbo LoRA

1.4K likes

Video generation with a synchronized soundtrack MiniMax-H3 — unquantized, split across two Spaces Joint video and soundtrack out of a single denoising pass, at bfloat16 with no quantization anywhere. This Space is the denoising half: the 61.73 GiB transformer and the two autoencoders. The 62.14 GiB Qwen3-VL conditioner runs in qwen3vl-conditioner, which this Space calls over the gradio API for every request. The weights are the public MiniMaxAI/MiniMax-H3 diffusers checkpoint. MiniMax-H3 is 195.9 GiB in bfloat16 and a ZeroGPU Space is evicted at 150 GB of storage. An unquantized single Space is therefore impossible, which is why quantized demos of it run NVFP4 or float8 weights. Cut the Mini...

Jev Decision Index

585 likes

Benchmarks and news on various repros of TypeSafe's Jev Who is rebuilding TypeSafe's Jev (System One / RLCD) in the open? This static Space opens on the Decision Index leaderboard; the News tab tracks the artifacts in one combined grid, color-coded by kind: Decoding: parallel constrained decoding on stock models (inference technique, no new weights) Diffusion: text diffusion models run in a "Jev mode" Trained: Jev-like scoring heads and fine-tunes, weights often on the Hub, promised models listed last Prior art: "this already exists" claims Explainers: architecture speculation, explainers, benchmarks and roundups Cards sort by a trending score: ♥ likes on X + 5 × GitHub stars + 8 × Hub likes...

Viggle Turbo for Qwen-Image-2.1

255 likes

6-step Qwen-Image-2.1, T2I + editing, vs-base comparison Viggle Turbo v0.3 — 6-step Qwen-Image-2.1 A distilled Qwen-Image-2.1 that generates and edits images in 6 steps with no classifier-free guidance, about 5× faster than the 40-step base model. On most prompts it is hard to tell apart from the base model; small, dense text and complicated edits (multi-reference composition, face swaps, identity-preserving edits) can still fall short of it. v0.3 (2026-09-29): at 6 steps, less grain than v0.2.1 and a little softer on fine texture. We think 6 steps is close to its capacity: every further gain we found cost something elsewhere. The new 9-step setting runs 7 turbo steps and lets the base model...

The ultimate guide to multi-harness RL

229 likes

Train open models with RL inside real agent harnesses A research article built with research-article-template. Source lives in FineEnvs under content/articles/multi-harness-rl/. | Path | What | | --- | --- | | app/src/content/article.mdx | Frontmatter and the chapter registry — the explicit import list is the running order | | app/src/content/chapters/ | One .mdx per section | | app/src/content/embeds/ | Standalone HTML/D3 visualizations, one file each | | app/src/content/assets/image/ | Images | | app/src/content/assets/data/ | Data files, served at /data/ | | app/src/content/bibliography.bib | References, cited as [@key] | From the repo root, over the Hub HTTP endpoint (no git remote, no n...

Model Pulse

183 likes

Hugging Face model download stats, history & trends Daily download history, likes and derivative families for every model on the Hugging Face Hub, going back to July 2024. Look up any model: ?model=org/name Compare up to five models on one chart See how much of a model's reach comes from its quantizations, fine-tunes, adapters and merges Galaxy: every model built on a base model, drawn as a galaxy (?view=galaxy) Wrapped: any author's last 12 months on the Hub in six cards (?view=wrapped) Weekly rankings: most downloaded, fastest growing, new breakouts, biggest families, organizations A README badge with monthly downloads and a sparkline, updated daily Data comes from daily snapshots of cfahl...

BFS Best Face Swap

172 likes

Head swap with the BFS LoRA on Qwen-Image-2.1 (+turbo) Head/face swap demo for the BFS bfsheadv1qwen2.1 LoRA on Qwen-Image-2.1, with an optional Viggle turbo 6-step mode. Example inputs are cropped from the example images in the BFS model repository (MIT license). Qwen-Image-2.1 and the Viggle turbo LoRA are under the Qwen Research License (non-commercial). Do not use or share results involving people who have not given consent. BFS Best Face Swap is a Hugging Face Space tagged with gradio, mcp-server, region:us. It has 172 likes on Hugging Face.