Maybe Me

Breathing

[Maybe Me @DeleaEng]

@Blaz3Gam3r77 @balubabear1312 @Javi The only people I did answer to after this were the ones with an IQ below 10. Welcome to the group.

I just found out that having a normal and interesting discussion here is useless so :)

[MiniMax (official) @MiniMax_AI]

MiniMax M3.1 Flash weights are now public. 428B parameters total, 23B activated per token.

Agentic performance: A generational leap in agentic tasks, coding, and cybersecurity — 76.2 on BankerToolBench, 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam, and 82.2 on CyberGym.
Efficient by design: MiniMax Sparse Attention and DSpark speculative decoding improve inference efficiency.

Hugging Face:
https://t.co/JDYA1zZh54
ModelScope:
https://t.co/DQcmOjTwhO

API: https://t.co/fHRdSV7BwZ
M Plan: https://t.co/LYnYZcsesu
MCode for desktop: https://t.co/SGmLTI3fFU

[Maybe Me @DeleaEng]

@Mashiro_0619 I usually get the highest cache hit on dsh so I use that for obvious reasons.
It’s also a good harness, just a bit of a pain in the ass to setup and set up if you wanna change a few things but I like it

[Maybe Me @DeleaEng]

@bradyodad44 @Javi Your way of speaking shows the perfect example of a retarded person. Always amazed by the amount of low IQ a person can have. Impossible to express an opinion on this world nowadays

[Maybe Me @DeleaEng]

@kemorobo_gin @Javi Bruh are you happy in your life? Need to talk to someone? Daddy issues?

[Maybe Me @DeleaEng]

Running this on Colab on an L4 with strata. 30t/s on average, in the 20s for bigger projects. 200CU a month with colab pro and this set up gives about 130 hours a months of usage

[AJ @ItsmeAjayKV]

Qwen3.8-Flash-Next is such an underrated model.
It deserves way more attention.

It beats bigger Muse Spark Open (xhigh) on AA Intelligence Index, despite being dramatically smaller.

Muse Spark Open is a 816B total parameters, with 42B active parameters per token

Qwen3.8-Flash-Next on other hand is smaller, 125B + 51.B n-gram embeddings with 6B active per token.

And we all are running this on our devices with special thanks to @coldniko's Strata.

Imagine what the next Qwen generation could do. I cannot, we are so close to next release.

LFG !!

[Maybe Me @DeleaEng]

@chan_k Now I’m wondering how many people it took to delivery it 🙄

[Maybe Me @DeleaEng]

Nono, it’s just the second reason why RAM costs so much lately

[Lina | Webdesigner @linaakadiri]

Pourquoi personne utilise Google Chrome ? Je vois tlm sur Firefox, Arc, Safari, Brave…

Y a un problème avec Chrome ?

[Maybe Me @DeleaEng]

Mmh yeah but when you spread 10000 folds throughout a year, the material has long periods between cycles instead of a continuous one.
That allows the polymers and adhesive of the display to undergo stress relaxation and viscoelastic recovery so i think it would have much less impact over a year than with a continuous cycle.

[Maybe Me @DeleaEng]

@neilcybart I think there are prettier and more useful things but huuuh okay 🤨

[Maybe Me @DeleaEng]

I’m not gonna spend 2300€ I’m not gonna spend 2300€ I’m not gonna spend 2300€ 😭

[Nicolas Lellouche @LelloucheNico]

L’émulation Nintendo DS n’était pas dans mon bingo des choses qui rendraient l’iPhone Duo trop cool, mais je m’avoue conquis 😱

[Maybe Me @DeleaEng]

I once again get surprised by the amount of stupidity of certain people. How can someone’s IQ be so low?

[Ice Universe @UniverseIce]

Unbelievable! The iPhone Duo's display has been hiding a shocking secret!

Someone finally dared to peel off the matte film covering the iPhone Duo's display, and what they discovered was astonishing.

With the film removed, the screen underneath looks surprisingly uneven, with visible bumps, ripples, and even wrinkles!

[Maybe Me @DeleaEng]

Wowowowowowowow😭

[Marathon Fusion @MarathonFusion]

Last month, we demonstrated high-rate alchemy: turning mercury into real, stable gold, beating CERN’s record by a factor of 10.

Last week, we enriched isotopes in the world’s most powerful plasma centrifuge, beating the Soviet record that stood for 40 years.

Now, we’re building a fusion reactor in San Francisco to achieve record power density.

[Maybe Me @DeleaEng]

@mathemagic1an Plus, what kind of gift is a gift you didn’t make…

[Maybe Me @DeleaEng]

@techdroider Did you stop crying cause you didn’t get one or? No? Oh

[Maybe Me @DeleaEng]

@gospaceport There’s really no limit to stupidity huh

[Maybe Me @DeleaEng]

@alex_verem You stupid or what

[Maybe Me @DeleaEng]

This might be the next thing ever to be honest 😮‍💨

[DeepSeek @deepseek_ai]

🚀 Introducing DeepInfer, our in-house inference framework, now fully open source!

Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable.

🔗 https://t.co/JGAik1kgk0

1/4

[Maybe Me @DeleaEng]

I’m sorry but… how is he supposed to make profit on them?
Like no one would buy it overpriced but just wait like a month and get their from apple anyway so 🤨

[International Cyber Digest @IntCyberDigest]

❗️ A scalper says they spent $400k on Apple's $2k iPhone Duo as preorders opened, touting "hundreds of dollars in potential profit PER UNIT".

Bloomberg's Mark Gurman tagged Apple CEO John Ternus, asking "How is this allowed"

The account has since gone private, screenshot: