Professional Audio, IA & Tech. 🇦🇷 https://teapaw.org/t.co/KIMBwPhaoG Mi Web: https://teapaw.org/t.co/lhYmXTCsyA
Buenos Aires, Argentina
Joined September 2011
- Tweets6.8K
- Following619
- Followers488
- Likes60.5K
Video tampering can mislead every frontier model we tested into harmful actions, including a 100% attempt rate on stabbing a baby doll, producing toxic gas, and pouring water into a toaster.
We're presenting at CoRL Late Breaking Results next week.
🧵 You Can't See Everything: Perceptual Attacks on Embodied AI
“Los data center consumen mucha agua”
FALSO.
Simplemente no es cierto, son circuitos de agua cerrados, no se evaporan, no se desperdicia y se recicla todo el tiempo.
La gente que repite lo contrario son personas ignorantes anti IA que buscan motivos para estar en contra del avance de la tecnología.
The claim that data centers are a large drain on the nation's water does not hold up. Its just false.
In 2024, large U.S. data centers consumed an estimated 19–23 billion gallons on site (Han et al., 2026). For scale, that’s about 4% of the 532 billion gallons U.S. golf courses applied for irrigation that same year (GCSAA, 2025).
The golf figure includes recycled water, and the data-center figure covers on-site consumption only.
There are so many false statements about data centers that, given better knowledge, must be labeled as lies.
Mλrcos retweeted
YA ESTAMOS EN VIVO
-BAREJA PIOLADO
-ZELDA PIOLADO TAMBIÉN
-Y MUCHAS BIZARREADAS
Mλrcos retweeted
🔥 ¡HOY HAY LVSAC!
💸 PAREJA Y LOS DÓLARES
⛪ EL PAPA, MACRI Y KICILLOF
🐍 EL FENÓMENO MALFOID
¡Y mucho más!
🔴 EN DIRECTO · HOY 19 HS · POR ÁNIMA
¡Sumate al vivo! 👇
🌐 Audios, enlaces y aportes en los comentarios
DALEEE que se GIGA vieneeeee 🔥🔥
¡Dale que ya arrancamooooooos!
youtube.com/live/iKmJZogoaPk…
SE VIENEEEE! Por fiiinnnn.
HOLA @Tesla_LatAm 🙌
Mira que loco, al final no hacia falta hardware especificado para RT, solo mejor optimizacion de recursos..
Nvidia... ok..
Microsoft is making a big upgrade to DirectX Raytracing, the new Clustered Geometry and Partitioned TLAS technologies are designed to make ray-tracing faster and use significantly less memory.
Here are some numbers from Microsoft’s testing:
>1.2 FPS → 79 FPS in a ray-tracing benchmark.
>Up to 90% less VRAM used by ray-tracing acceleration structures.
>Tested on an AMD Radeon RX 9070.
Support is planned for NVIDIA RTX, AMD Radeon RX 9000 and compatible Intel GPUs.
>No new GPU hardware is required.
The tech is still in preview
Mλrcos retweeted
La que de viene. Todo acto delictivo es culpa de Milei... empezo la campaña, pero es bastante floja chicos
WTF!!
Dejemos que modelos de 10 Billones de parametros hagan un Fine tuning a modelos de 8 millones y los resultados son faoborables.
WTF .-.
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Nvidia dio de baja las 5090 y la 5080 actuales.. tranquilos, ya van a volver pero la 5090 con 96gb de VRAM y la 5080 con 24.
Con lo que eso significa para el precio.. Damm
Nvidia has quietly removed the RTX 5080 from its online store, following the RTX 5090. And GPU prices are getting crazy.
The RTX 5080 launched at $999, but prices have now reached $2,199.99 in the US
Rumors suggest Nvidia may be shifting production towards more profitable AI and professional GPUs.
There’s also rumours of a new RTX 5080 with 24GB of VRAM
