Extremely excited about our internal evals! Abacus AI Desktop is topping another very important bench We will announce shortly
GENERATIVE AI
-
Scaling improvements will continue despite missing critical elements
By
–
One point I made that didn’t come across:
– Scaling the current thing will keep leading to improvements. In particular, it won’t stall.
– But something important will continue to be missing. x.com/slow_developer… -
Apple Overtakes Samsung; MIT Studies AI Job Impact
By
–
Top stories in tech today: – Apple set to overtake Samsung for top spot
– MIT says AI is ready for 12% of jobs
– Alibaba launches Meta-rivaling AI glasses
– China rescues astronauts stranded on space station
– Quick hits on other tech news Read more: https://
tech.therundown.ai/p/apple-takes-
the-crown-from-samsung
… -
AI Inference Costs Collapse 99.9% Reshaping Economy
By
–
Three years ago, a million tokens of AI inference cost $60. Today? Six cents.
— Nina Schick (@NinaDSchick) 28 novembre 2025
A 99.9% cost collapse. When something this powerful becomes this cheap, it doesn’t stay confined to research labs, It is able to flood the economy.
AI is now diffusing faster than any technology in… pic.twitter.com/pObTHhuBC8Three years ago, a million tokens of AI inference cost $60. Today? Six cents. A 99.9% cost collapse. When something this powerful becomes this cheap, it doesn’t stay confined to research labs, It is able to flood the economy. AI is now diffusing faster than any technology in
-
AI Exceeds Humans in Technical Tasks, Humans Lead in Multimodal Reasoning
By
–
AI now exceeds human performance in most technical tasks — but humans still lead in multimodal reasoning The next frontier isn’t about replacing humans, but amplifying how we think across data, images & language. Source: Stanford AI Index 2025
#AI #Innovation #FutureOfWork -
Claude Scientific Skills: Open-Source AI Research Assistant Platform
By
–
Claude Scientific Skills. Turn Claude into your AI research assistant capable of executing complex multi-step scientific workflows across maths, biology, chemistry, medicine, and beyond. 100% open-source (123+ skills).
-
V2V Feature Anticipation in Next Generation AI Model
By
–
Idk why they haven’t released v2v tbh; hopefully in veo 4
-
Veo Preferred Over Flow for Text-to-Video Generation
By
–
Def prefer Veo for text/img to video. Flow doesn’t have video to video yet.
-
Technical discussion on embeddings in generative AI
By
–
Mais tu considère les embeddings comme de l’IA Gen ? Pour moins c’est utile aux LLMs, mais les embeddings ont en faisait avant
-
Gemini 3.0 vs ChatGPT-5.1 vs Claude 4.5 Comparison
By
–
Gemini 3.0 Pro vs ChatGPT-5.1 Thinking vs Claude 4.5 Opus I tested all these powerful LLMs using critical prompts. The results shocked me. (Demos + prompts)