Everywear

Local-first AI OS

[Everywear @everywearos]

@alexutopia https://t.co/uMVQOgdsYX

[Everywear @everywearos]

@alexutopia https://t.co/Qqn9jjej4Y :D

[Alex Utopia @alexutopia]

An experiment in agentic filmmaking.

I gave an agent 13 video clips and one song.
Then let it take the director's chair.

Editing: GPT-6 Astra
Video: Grok Imagine
Music: Suno v6

[Everywear @everywearos]

https://t.co/k5Shjepol2 beta 0.3 live now, trial keys come shortly if you want to try before you buy. #wardogs #mw4 #everywear #mymait #everywearos

[Everywear @everywearos]

Wing Mait™ | Your AI squadmate on comms.

It sees what you see and says it out loud. No injection, no memory reads.

Windows beta: https://t.co/QcOPfYg5LT
Made by Metafintek • https://t.co/HVfV2Rg2r4
#wardogs #wingmait

[David Hendrickson @TeksEdge]

How did I miss this?! Bonsai isn’t just an LLM family.

Back in May, PrismML released Bonsai Image 4B, a crazy low-bit version of FLUX.2 Klein 4B designed to run locally on 🍎 iPhones.

Look at this
🎨 FLUX.2 Klein 4B transformer
💾 FP16: 7.75GB
🌳 Ternary Bonsai: 1.21GB

And the entire Apple deployment payload, including its compressed text encoder + VAE, is only:

🔥 3.88GB

📱 iPhone 17 Pro Max
🧠 A19 Pro / 12GB unified memory
🖼️ 512×512
⚡ 9.4 sec/image
📲 MLX Swift
☁️ No cloud
💾 Bomsai Studio (App Store)

M4 Pro: ~5.8 sec/image.

No cloud.

The ~4B diffusion transformer is paired with a 4-bit Qwen3-4B text encoder, which gets unloaded after the prompt is encoded to save memory.

And there’s also
🍎 Mac / iPhone / iPad support
🟢 low-bit Gemlite builds for Nvidia GPUs
🔓 Apache 2.0

This is completely separate from the Bonsai 2 27B LLM I’ve been posting about.

PrismML took a 7.75GB FLUX transformer and crunched it down to 1.21G and put image generation on an iPhone. 👀

And somehow I missed this for 4 months.

[Ｂ４ＳＩＣ @B4SICAI]

Lagi siapin semuanya buat Beta hari Senin 🚀
WingMait, teman gaming pribadi yang bantu Anda main lebih cerdas, lebih sadar situasi, dan lebih siap di medan tempur.
🎮 Dibangun untuk gamer Indonesia dan APAC
🌐 https://t.co/13cGvFWq4l
#Indonesia #APAC #WarDogs #WingMait

[Ｂ４ＳＩＣ @B4SICAI]

准备好了，周一 Beta 见！🚀

WingMait，你的智能游戏战友。
帮你更快掌握战局，更清楚地了解战场动态，在关键时刻做出更好的判断。
🎮 为中国及亚太地区玩家打造

https://t.co/FIwcg0DXfj

#中国 #亚太 #WarDogs #WingMait #PTMetaFinTek

[Ｂ４ＳＩＣ @B4SICAI]

तैयार हो जाइए, सोमवार को Beta शुरू हो रहा है 🚀

WingMait आपका स्मार्ट gaming साथी है, जो आपको battle जल्दी समझने और बेहतर फैसले लेने में मदद करता है।

🎮 भारत और APAC के gamers के लिए

https://t.co/qc8aTKTZcF
#India #APAC #WarDogs #WingMait

[Ｂ４ＳＩＣ @B4SICAI]

เตรียมตัวให้พร้อม เจอกันใน Beta วันจันทร์นี้ 🚀

WingMait เพื่อนร่วมทีม AI ที่จะช่วยให้คุณอ่านเกมไวขึ้น จับสถานการณ์ได้ดีขึ้น และตัดสินใจได้แม่นกว่าเดิม

สร้างมาเพื่อเกมเมอร์ไทยและสายลุยทั่ว APAC 🎮

https://t.co/OOirlyXLEf

#Thailand #APAC #WarDogs #WingMait #PTMetaFinTek

[Everywear @everywearos]

Really Fun product and a great application of everywear

[Ｂ４ＳＩＣ @B4SICAI]

Guys for a limited time I shall be back streaming #wardogs in conjunction with @everywearos showcasing https://t.co/x6RpJp9TSx
Welcome. Enjoy the war.
https://t.co/4lboPcu5az

[Eyisha Zyer @eyishazyer]

"The Tubes": a dystopia where the poor survive inside glass tubes while the rich watch. (yes it's AI)

[Ｂ４ＳＩＣ @B4SICAI]

Literally the most exciting thing in AI.

[Everywear @everywearos]

@pankajkumar_dev @aisolram After working with both Claude max 200.and codex max for the last month on https://t.co/oduVo1rEpB Claude max finally admitted it wasn't good enough. Downgraded the mobthly max to yearly pro. Codex + qwen 36b local win.

[Everywear @everywearos]

@Michal_Stan @pankajkumar_dev You think engineers built it. How cute.

[Everywear @everywearos]

@pankajkumar_dev They are all gone when subqad +jepa hit their stride in 18 months or so.

[Everywear @everywearos]

Amazing!

[Strands @strandsnation]

Beta is live right now! https://t.co/xBq5v8jV4I

[TechHalla @techhalla]

China is already making this a reality.

You can now preserve a late family member or pet like an AI Skill so they can stay by your side always.

Cringe or cool? 🫣

[Everywear @everywearos]

@sama Recently became a little dumbed down im afraid. semi psycophantic, a distinct lack of challenge to thinking. WAy too much reiteration of what i just told it. IF i igive it information i want it to extrapolate and add to the conversation, not waste tokens reiterating what i said

[Everywear @everywearos]

@sama When you use Gpt so you use the same cloud as everyone else or do you have a closed "offline" iteration?