DeepSeek now also undermines the business case for scaling. OpenAI has been burning through staggering sums of cash to keep up its scaling paradigm and has yet to figure out how to balance its checkbooks – and it turns out it didn’t need to spend so much cash. 18/
LLMS
-
DeepSeek challenges OpenAI’s narrative on AI development trade-offs
By
–
But DeepSeek shows – in the same moment as the Stargate announcement – that the trade off that OpenAI & co frame as wholly necessary is actually not. 15/
-
DeepSeek-R1-distill now on Groq
By
–
NEW : DeepSeek-R1-distill model is now available on groq It performs less than a full o1 but is much faster. * keep in mind that this is not a DeepSeek R1
-
Text Compression Limits in Large Language Models
By
–
It’s hard to compress text deterministically beyond a certain point but LLMs seem quite far from that point. I’d certainly prefer such a model…
-
Correction: Model Performance Comparison O1 Versus O3
By
–
Petite coquille : plus performant que o1, pas o3. Ce qui ne change absolument rien au fond du désastre.
-
Deepseek R1 challenges GPT o3 in AI competition between US and China
By
–
Le monde de la tech est en ébullition depuis le lancement de l’IA chinoise Deepseek R1, visiblement plus performante et légère que GPT o3. Le choc des titans a commencé, entre d’une part une Amérique qui va déréguler à tout-va, et d’autre part une Chine qui va tenter d’imposer
-
New AI Canvas Features for Real-time Code Rendering and Testing
By
–
The new Canvas: Renders HTML/React code in real-time Includes a built-in Python emulator for immediate results (yes, even visual ones!) Lets you highlight and tweak specific output directly It’s intuitive, seamless, and ready to save devs HOURS of testing and
-
Supervised Fine-Tuning vs Reinforcement Learning: A Learning Analogy
By
–
“Imagine teaching a child to ride a bike. You could give them a detailed manual (Supervised Fine Tuning), but they'll likely learn better by trying it themselves (Reinforcement Learning), falling, getting up, & gradually improving.” – @McDonaghMatthew ELI5 on DeepSeek, link
-
Comparing grok-2-latest vs grok-3 prompts
By
–
grok-2-latest 🤖 vs grok-3 🤖
— 🚨 AI News | TestingCatalog (@testingcatalog) 26 janvier 2025
What would you ask to impersonate who is who?
Drop your prompts below 👀 pic.twitter.com/JmgfREFPEVgrok-2-latest vs grok-3 What would you ask to impersonate who is who? Drop your prompts below
