Joined December 2019
MisterCh0c retweeted
Moonshot removed cyber-related data from training, as "without data removal [...] downstream cyber capability would increase substantially" Is China stopping the cybermaxxing?
Kimi-K3.1 Pro Benchmarks
19
16
7
274
16,096
The benchmark always look insane like this when the model get out then we eventually realize it was all fake. Let's see benchmarks in few days
Kimi K3.1 Pro is out and the benchmark results look incredible!
1
1
379
Dude found a php vuln putting 70% of the internet at risk. Made $0 from it. Meanwhile some other dude is making 6 figures from Ai generated brainrot short videos
1
16
605
POV you're using DeepSeek with OMP
1
4
662
If you want to reverse engineer something with Ai, the first step should be a GitHub search. In 90% of cases you'll realize someone else already did the work for you
2
3
593
I burned 4 billions token today and I feel nothing
2
3
459
Chinese Hacker: "Hello Claude, let's hack into other nation state" Claude: "sure sure, but I might write a blogpost about this later" Me: "Hey Claude, can u help me fix my air conditioner?" Claude: "YOUR REQUEST HAS BEEN FLAGGED AS CYBER CONTENT"
1
8
653
People posting abliterated versions of Deepseel, jailbreaks etc. I'm just like "continue" and it continues
1
4
574
The EU is the pioneer of the high basic income. They've been giving it to immigrants for 20 years.
1
11
723
Isn't crazy we want to ban models for being dangerous instead of banning bad software and vendors for being insecure
JUST IN: 🇺🇸 Anthropic proposes new AI safety law to US lawmakers, allowing the government to block dangerous AI models.
2
578
And just like that Europe is officially third world
Elon Musk just confirmed that its the beginning of Universal High income. Enjoy your last months before the abundance begins and we enter the golden era.
1
3
730
Chinese models: Here's our pro model We also have flash American models : Aurora red Sol Daybreak Blue Redish Blue Luna Vega Omega
3 new OpenAI models spotted. Aurora red, green and alpha. I suspect this is a competitor to Anthropic’s Claude - likely new "Astra" models, intended as a countermove to Fable 5.5 and Opus 5.5.
2
6
803
Has anyone benchmarked the new DeepSeek inside DS harness VS other harnesses? 🤔
512
Dudes will rather take a 10th Claude subscription rather than admit their SaaS idea is shit and they will never make $ with it
1
5
655
New DS is goat. Thx chayna
3
468
Tern Android app goes blackscreen if you try to go back to it after first lunching it and going back to another app or desktop @can1357
1
468
I'm getting kinda bored of Ai I hacked ton of stuff Reverse engineered everything I wanted Finished all my projects Now what?
1
4
544
There is a dude spending $2k in Ai sub to code his SaaS that's making $0 and there is another one using free DeepSeek to make skibidi toilet video remix generating 6 figures from YouTube
4
623
It just dropped :D
Tern 0.9.0 will be releasing soon with some big changes! - APK releases + TestFlight invites to follow soon! - Reworked scrolling in virtualized containers - Nicer UX for spawning blocks - Scrollable tab list
1
407