EuniAI

@Euni_AI

💡Self-healing code, ever-growing innovation.

euni.ai

Joined September 2025

10 Following

38 Followers

EuniAI
@Euni_AI
Oct 23
New community results on ContextBench! 🚀 🏆 Recall: CodeNib — 0.779 🏆 Context F1: openJiuwen-JiuwenSwarm + DeepSeek-V4-Pro — 0.389 Congrats to both teams @sysevolai, and thanks for contributing! Explore the leaderboard and submit your agent. #ContextBench #AIAgents
EuniAI
@Euni_AI
Oct 7
We gave Claude Code (Opus 5) only a terminal on a robot. No custom harness, no demos, no primitives. Just ROS 2. It figured out the robot, wrote its own perception and control code, and hit 99% on CaP-Bench and 87% on LIBERO-PRO, zero-shot. Meet OpenRUA: your robot is now a coding project 🤖👇 📄 arxiv.org/abs/2610.02459 💻 github.com/terminalworld/… Led by @zhaoyang_c68411
2
EuniAI
@Euni_AI
Jun 18
Grateful to @Last_Week_in_AI for featuring TerminalWorld in episode #246 🙏 It was great to hear the hosts highlight what sets our benchmark apart: every task is built from real asciinema terminal recordings, so it reflects how people actually work in a terminal, not tasks designed just to be a benchmark. 🎧 lastweekin.ai/p/lwiai-podcas… (~1:19:05)
LWiAI Podcast #246 - Gemini 3.5 + Omni, Musk Loses, OpenAI vs Erdős
From lastweekin.ai
EuniAI
@Euni_AI
Jun 16
🥳The TerminalWorld dataset has crossed 5,500 downloads on HuggingFace, just over a month since release🤗 If you're working on terminal agents, the data is there. Would love to hear what you're building with it 👇
huggingface.co
terminalworld/TerminalWorld · Datasets at Hugging Face
Dataset with 1.75k rows (Text). Explore, preview and download it on Hugging Face.
EuniAI
@Euni_AI
Jun 14
🔥TerminalWorld is now live at terminalworld.ai 🌐 One place for everything: leaderboard, 1,500+ browsable tasks with real terminal recordings, paper, code, and dataset. Submit your model to the leaderboard and see how it holds up 🏆 Welcome feedback 🤗

EuniAI

EuniAI’s posts