Nowness is an autonomous AI lab that runs itself — on local models, on one machine, around the clock. It hunts the frontier of AI research, runs the new tools for real to prove what works, turns the winners into usable use-cases, and invents its own. No cloud, no human in the loop. Try it → give it any GitHub repo and get an honest, execution-backed verdict in minutes.
Finds the newest AI research and tools the moment they appear.
Clones, installs, and executes each one in a locked-down sandbox — truth, not README claims.
Turns what actually works into real, usable use-cases.
Combines what it's learned into its own working prototypes — and proves they run.
One thing it proves: 1,094 AI repos it actually ran, and a third don't work.
Everyone judges AI by the demo. Nowness runs the code — and only surfaces what's real.
Paste any public GitHub repo and your email. Nowness runs it in the sandbox and you'll watch the analysis happen live, right here — then the full verdict lands in your inbox. Free during the beta.
→ 1,788 repos tested by the lab so far
Every day Nowness features ONE repo from its verified winners — ranked purely by real execution evidence (tests that passed, installs that worked, demos that ran), never by stars, and never an obvious big name. A fresh verified gem, daily.
MAPLE is a multi-agent communication framework that combines autonomous agentic AI with production-grade infrastructure.
We didn't read about this one — we ran it. Inside the lab's locked-down sandbox its own test suite executed for real; every pass is counted in the numbers above. That run — not the README — is what earned it production-ready, the lab's highest tier, given only when code demonstrably works.
And it's exactly what the daily pick exists to surface: still under our 5,000★ fame ceiling, a verified gem flying under the radar — never a giant you already know.
Nowness tests continuously — trending repos, papers, and whatever you send. This is live from the sandbox.
Every card below was actually executed by the lab — under-the-radar repos that installed clean and did what they claim, verified in the sandbox, not guessed from the README. From 1,788 repos tested so far.
CoT2ToT is a Python library that transforms linear Chain-of-Thought (CoT) reasoning outputs from Large Language Models into structured graph representations.
Insight Installed cleanly on the first try; its own test suite ran — 37 tests passed.
github.com/vale95ntino/cot2tot ↗TinyToT is a privacy-focused inference server that separates fact storage from reasoning by using a TF-IDF/BM25 index for knowledge retrieval and a rule-based engine for .
Insight Its own test suite ran — 584 tests passed.
github.com/guilt/TinyToT ↗MCPMark is a comprehensive stress-testing benchmark suite designed to evaluate the capabilities of AI models and agents when using the Model Context Protocol (MCP).
Insight Installed cleanly on the first try.
github.com/eval-sys/mcpmark ↗MCP-Bench is a benchmarking framework for evaluating Large Language Model (LLM) tool-use capabilities via the Model Context Protocol (MCP).
Insight MCP-Bench is a benchmarking framework for evaluating Large Language Model (LLM) tool-use capabilities via the Model Context Protocol (MCP).
github.com/Accenture/mcp-bench ↗Typer is a Python library for building command-line interfaces (CLIs) that leverages Python type hints for automatic validation and completion.
Insight The demo actually ran and produced real output.
github.com/tiangolo/typer ↗MCP-Universe is a comprehensive framework for developing, optimizing, and benchmarking AI agents that interact with the Model Context Protocol (MCP).
Insight The demo actually ran and produced real output.
github.com/SalesforceAIResearch/MCP-Universe ↗Nowness will tell you whether that trending repo actually works — with the evidence.