Llama 4 Scout delivers state-of-the-art performance for its class enabled by continued “mid-training” with new training recipes using specialized datasets enhancing model quality and unlocking a 10M token input context length.
RESEARCH
-

Llama 4 Scout and Maverick Now Available on Poe
By
–
The first versions of Llama 4 Scout and Maverick are now available on Poe, thanks to @FireworksAI_HQ
! -
Google’s Integrated AI Stack: Hardware to Cloud Services
By
–
Google has the entire AI stack: hardware (TPUs), data centers, large web datasets, training infrastructure, research, and cloud services to serve models. The ability to optimize across all parts of the stack end-to-end has incredible advantages.
-
109B total params, 17B active, 16 experts
By
–
109B 17B active params x 16 experts, 109B total params
-

Meta Introduces Llama 4 Scout and Maverick Multimodal AI Models
By
–
Today is the start of a new era of natively multimodal AI innovation. Today, we’re introducing the first Llama 4 models: Llama 4 Scout and Llama 4 Maverick — our most advanced models yet and the best in their class for multimodality. Llama 4 Scout
• 17B-active-parameter model -
Major AI Demo Completed, Public Announcement Coming This Month
By
–
alright — demo done today. I think it went well! We’ll publicly announce sometime this month. I’m off twitter for a bit (health stuff). Catch you public demo time!
-

LLMs Enable New Approaches to Social Science Research
By
–
LLMs are remarkably accurate at replicating some types of human behavior. This opens the door to a whole new way of doing social science research.
-
Continuous Improvement: Study and Practice of Better Methods
By
–
Continuous improvement is the study and practice of how to do things better and how to make things better.
-

LLM ML Job Interviews Process Fall 2024
By
–
LLM (ML) Job Interviews (Fall 2024) – Process • Mimansa Jaiswal https://
buff.ly/kkVLafo
#AI #MachineLearning #DeepLearning #LLMs #DataScience -

NoProp: Training Neural Networks Without Back-propagation
By
–
NoProp: Training Neural Networks without Back-propagation or Forward-propagation Qinyu Li, Yee Whye Teh, Razvan Pascanu: https://
arxiv.org/abs/2503.24322 #ArtificialIntelligence #DeepLearning #MachineLearning
