Even a year or two ago with a whole bunch of big-but-weak models, routers weren’t too helpful… but things are a-changin’!
LLMS
-
Llama-3.1 Models Ranked on Chatbot Arena
By
–
We finally have an Chatbot Arena ranking for Llama-3.1 models! 405B Model ranks 5th overall, in front of GPT-4-turbo! But behind GPT-4o, Claude-3.5 Sonnet and Gemini-advanced. 70B Model climbs up to 9th rank ! From 1206 1244. 8B Model improves from 1152 1170.
-
Exploring Classical Regression Capabilities of Large Language Models
By
–
GitHub – AlexIoannides/llm-regression: Exploring the classical regression capabilities of LLMs. https://
bit.ly/3zF6TkT
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
Embedding Models: Architecture to Implementation Course
By
–
Learn how embedding models are built, trained, and used in semantic search systems in this new short course, Embedding Models: From Architecture to Implementation, created with @vectara and taught by @ofermend.
— Andrew Ng (@AndrewYNg) 31 juillet 2024
Many LLM apps use a single embedding model for both questions and… pic.twitter.com/LjiVgPECqELearn how embedding models are built, trained, and used in semantic search systems in this new short course, Embedding Models: From Architecture to Implementation, created with @vectara and taught by @ofermend
. Many LLM apps use a single embedding model for both questions and -
Must-see AI tools and releases from past month
By
–
As the news cycle takes over the feed, I feel like I must be missing some exciting new AI stuff. What are your favorite must-see AI tools, releases, and news from this past month?
-
AI Research Graph Featured in German Newspaper
By
–
Haha cool to see my graph in a German newspaper (Frankfurter Allgemeine Zeitung) 🙂 (But R.I.P. Falcon 180B)
-
New GPT-4o Voice imitates a sports commentator
By
–
New GPT-4o Voice performs with intense intonation and emotion as it imitates a soccer commentator:
— AI Breakfast (@AiBreakfast) 31 juillet 2024
(Advanced voice model was released to select users last night)pic.twitter.com/PTHH7CqtlWNew GPT-4o Voice performs with intense intonation and emotion as it imitates a soccer commentator: (Advanced voice model was released to select users last night)
-
Google Launches Gemini 1.5 Flash Free AI Model
By
–
Google has launched Gemini 1.5 Flash, an unpaid version of Gemini, for faster and more helpful responses. They have also introduced a new feature to further address hallucinations and expanded the Gemini for Teens experience and mobile app to more places. Users can now do things
-
Model routing implementation best practices
By
–
Looks like model routing done right! https://t.co/qsnjxkGMJs
— Yohei (@yoheinakajima) 31 juillet 2024Looks like model routing done right!
-
Regrets and Regression in Machine Learning Systems
By
–
Regrets and Regression https://
bit.ly/3zNa1uQ
#AI #MachineLearning #DeepLearning #LLMs #DataScience