The open source cowork desktop agent. Github: https://nitter.cf/t.co/Qcyzi1Roxp
Joined September 2024
- Tweets223
- Following8
- Followers3.2K
- Likes233
Pinned Tweet
@Zai_org GLM 5.2 is now live on Eigent from day 0!
threw a real long-horizon task: research 30 companies across 6 sectors of the AI infrastructure stack, structure it into JSON, then build an interactive HTML report. same prompt, 5.1 vs 5.2. where 5.2 pulls ahead:
→ plans deeper and verifies harder.
→ researches wider. Covers 2x the ground.
→ ships a more complete, interactive report.
with a 1M context window, it holds the whole task from research to final deliverable. no dropped threads, no half-finished output.
long tasks are where the difference shows. try it on open source coworker Eigent. 🦾
Eigent v1.0.5 is live! Your agent workspace is easier to recover, control, and review.
What’s new:
🔄 More reliable Session recovery and Task history
📋 A reorderable Task queue
🖥️ Background process controls and better file and browser previews
🧩 Space resources and model defaults
✨ Meta Model API and Muse Spark support
🌿 An opt-in preview of parallel local Sessions for configured deployments
⚙️ Fixes across Settings, Git paths, Windows startup, and credentials
Give v1.0.5 a try and let us know what you think! Feedback, bug reports, and ideas are welcome 🙌
eigent.ai/blog/eigent-v1-0-5…
Gemini 3.8 now on Eigent! ⚡️⚡️⚡️
We handed it a real finance analyst's job: value Equinix vs Digital Realty, Damodaran-style, straight from SEC filings.
A team of agents researched, checked FX, and built the DCF, then handed back an Excel file you can actually poke at.
Click a number → see the filing it came from. Change an assumption → watch the valuation shift live.
Gemini 3.8 is now supported on Eigent via BYOK and cloud API. Try it today！
Eigent v1.0.3 is live! Your agent workspace is now more durable, reviewable, and reusable!
What’s new:
🔄 Durable Tasks That Can Resume Safely
🌿 Git-Backed Spaces with Task-Scoped Change Review
🖥️ Browser, Files, and a Real Terminal Inside the Session
📄 Generated Files That Are Easier to Open and Review
📦 Workspace Bundles: Reuse the Whole Working Setup
🧠 Scoped Memory Without Rewriting Task History
🛡️ Safer Approvals, Recovery, and Workspace Operations
✨ Ant Ling Support and Focused UX Fixes
🔁 Compatibility for Existing Work
Give Eigent v1.0.3 a try and let us know what you think! Feedback, bug reports, and ideas are all welcome 🙌
eigent.ai/blog/eigent-v1-0-3…
Eigent AI retweeted
I’ve been fortunate to work with the amazing @radixark team @zz30gs @ShiDong14 , yueming, jiajun post-training DeepSeek V4 Flash, Inkling Small, and GLM-5.2 LoRA on terminal environment we produced at @Eigent_AI .
Miles is my top recommendation for RL. Congrats on the launch!
Today we're launching Miles v0.1, an open-source RL framework for LLMs and multimodal models.
RL training is easy to start and hard to debug. Miles helps you ensure your run is correct, use hardware efficiently, and keep RL running at scale.
Over the past 9 months, 72 contributors have landed 1,326 commits, 85 GPU E2E CI tests, battle-testing Miles on frontier open models like Kimi K3, DeepSeek V4, Qwen 3.8, GLM 5.2, Inkling, MiniMax H3, etc.
Miles powers frontier-model development and production RL workloads at @humansand, @periodiclabs, @modal, @DecagonAI, @Eigent_AI, @nebiusai, @IBM and more, on both @NVIDIAAI and @AIatAMD hardware.
Here is what we built, and why teams picked Miles🧵
Congrats to @radixark on Miles v0.1! We are using Miles to train RL agents for terminal coding and ML engineering across thousands of environments, with models like DeepSeek V4 Flash and GLM 5.2 and SGLang as our rollout engine. Efficient async RL makes scaling easy, and the team behind it is incredibly supportive!
Today we're launching Miles v0.1, an open-source RL framework for LLMs and multimodal models.
RL training is easy to start and hard to debug. Miles helps you ensure your run is correct, use hardware efficiently, and keep RL running at scale.
Over the past 9 months, 72 contributors have landed 1,326 commits, 85 GPU E2E CI tests, battle-testing Miles on frontier open models like Kimi K3, DeepSeek V4, Qwen 3.8, GLM 5.2, Inkling, MiniMax H3, etc.
Miles powers frontier-model development and production RL workloads at @humansand, @periodiclabs, @modal, @DecagonAI, @Eigent_AI, @nebiusai, @IBM and more, on both @NVIDIAAI and @AIatAMD hardware.
Here is what we built, and why teams picked Miles🧵
Video in, print-ready 3D model out in 3 minutes. Gemini 3.7 Flash is live on Eigent!
Gemini 3.7 Flash now live on Eigent on both byok and cloud api ⚡️
Lightning speed meets unbelievable precision:
> Video-to-3D understanding
> Ultra-detailed, articulated CAD/mesh models (.glb)
> Full PBR materials & emissive lighting
From video reference to production-ready 3D in 3 mins! We also did a comparison with 3.6 Flash it's 2x faster and delivers far richer, more accurate details! 🚀
Download Eigent with free credits to build your own 3D models today!
@deepseek_ai V4 Flash now available on Eigent! ⚡
Fire up Eigent Single Agent Mode with DeepSeek V4 Flash and watch your task get executed instantly!
We demoed V4 Pro on Eigent two months ago to ship entire monthly dev report: GitHub PRs → Word doc → Slack message → sent to product-release.
Full walkthrough in the thread 🧵👇
👉 Sign up today for free credits and start building for free.
🚀 DeepSeek-V4-Flash Official API is now LIVE in public beta!
🔷 We’ve massively upgraded its Agent capabilities—benchmark scores are now far surpassing the V4-Pro-Preview. Check out the massive performance leap below! 👇
🔷 The official V4-Flash now natively supports the Responses API format and is fully adapted for Codex!
Check out the configuration details in our official API docs: api-docs.deepseek.com/quick_…
👉 Automate Monthly Dev Reports with DeepSeek via @ollama
eigent.ai/use-cases/monthly-…
Eigent v1.0.2 is here.
One of our biggest updates yet, bringing Connector Gateway, a unified in-session preview, deeper BYOK controls, more reliable long-running workflows, and a smoother experience across chat and desktop.
🔌Connector Gateway to bring your apps directly to your agents
💻 One in-session preview panel for every kind of work
🎛️ More control with BYOK model parameters and session-level model choice
📎 Paste images directly into chat
🔎 Clearer work logs with typed workflow events
🛡️ More reliable long-running workflows
✨ Plus improvements across chat, desktop, sign-in, OAuth, Windows, Docker, and more
A smoother, more connected, and more reliable Eigent experience.
Dive into all the PRs and technical details in our full release blog 📷 eigent.ai/blog/eigent-v1-0-2…
Gemini 3.6 Flash and 3.5 Flash-Lite are now live on Eigent ⚡️ @GoogleDeepMind
In just 50 seconds, Gemini 3.6 Flash turned one prompt into a fully playable 3D browser game where you control a moving hole, swallow houses and city objects, grow bigger, and rack up points as you explore the city. 🌆
Sign up for Eigent to get free credits and start building with the latest models 🚀
We’re rolling out three new models to make AI agents faster, smarter, and cheaper at scale:
🔵 Gemini 3.6 Flash: It uses fewer tokens than 3.5 Flash to deliver higher quality work at the exact same cost.
🔵 Gemini 3.5 Flash-Lite: A fast, cost-effective option for everyday tasks like processing documents and agentic search.
🔵 Gemini 3.5 Flash Cyber: A cybersecurity model built to find and patch critical software vulnerabilities.
GPT 5.6 is now live on Eigent, it is surprisingly good at turning scientific ideas into practical plans.
We asked Eigent to explore The Lunar Garden, an idea about growing plants in lunar soil. 🌕 It researched credible science, designed a plant experiment, planned daily observations, and built a 30 day growth simulation with an interactive story concept.
Download Eigent and try GPT 5.6 for free!
Gave GPT 5.6 on Eigent a mission: turn a scientific idea into a working plan.
We asked Eigent to explore The Lunar Garden — an idea about growing plants in lunar soil. It researched credible science, designed a plant experiment, planned daily observations, and built a 30 day growth simulation with an interactive story concept.
What usually stays a rough hypothesis becomes a structured, testable plan you can actually follow.
Built with GPT 5.6 on Eigent, free for new users to try!
Gave Grok 4.5 on Eigent Cloud a mission: make the Big Falcon Ship architecture feel explorable 🚀
it turned a complex spacecraft cutaway into an interactive explorer, mapping the ship into propulsion, landing systems, aero surfaces, life support, guidance, bridge controls, cargo, and crew spaces.
scroll through the ship and each system comes into focus. click any module to see what it does.
what usually feels like a dense aerospace diagram becomes a cinematic walkthrough you can understand at a glance. 👀
built with grok 4.5 on Eigent cloud, free for new users to try!
Announcing Grok 4.5, our first model trained specifically for coding and agents. It was trained with Cursor and offers frontier intelligence at leading speeds and cost efficiency.
x.ai/news/grok-4-5