Woo hoo, my friend @chrisdumler pointed out to me that Sonnet 3.5 knows about BabyAGI! (GPT-4 and 4o do not) It knows about me too!
LLMS
-
Questioning model selection: GPT-4 versus GPT-4o
By
–
Any insights why gpt4 and not 4o? Or is it 4o in fact?
-
Using custom prompts in Perplexity Collections
By
–
Perplexity “projects”? Not sure how new is this but you can add a custom prompt to your Collections on Perplexity. All new threads started from such collection will have it applied. I submitted it as a feature request 2 weeks ago but now not 100% if it was already there
-
Grok’s competitive advantage through X data integration
By
–
For sure! But I am afraid that X might be too pricy to integrate with. Access to X live data is a big differentiation point for Grok Not sure how can it be played out
-
LLMs Combined with Monte Carlo Tree Search Beat GPT-4
By
–
Combining LLMs with other techniques yields promising results! This paper (link in alt) shows that combining LLMs with Monte Carlo Tree Search enhances performance in complex mathematical reasoning tasks. With this, a small 8b Llama-3 was able to beat GPT-4!
-
Poe Introduces Previews: Web Apps Generated by AI in Chats
By
–
We’re excited to introduce Previews, a new feature that lets you see and interact with web applications generated directly in chats on Poe. Previews works particularly well with LLMs that excel at coding, including Claude 3.5 Sonnet, GPT-4o, and Gemini 1.5 Pro. (1/5) pic.twitter.com/aOqrr4Jv5M
— Poe (@poe_platform) 8 juillet 2024We’re excited to introduce Previews, a new feature that lets you see and interact with web applications generated directly in chats on Poe. Previews works particularly well with LLMs that excel at coding, including Claude 3.5 Sonnet, GPT-4o, and Gemini 1.5 Pro. (1/5)
-
Yi Integrated with LlamaIndex for Enhanced RAG Applications
By
–
🎉Exciting News: we've integrated #Yi with @llama_index for more efficient retrieval and indexing capabilities, helping you build smarter and faster RAG apps with ease!
— Yi-01.AI (@01AI_Yi) 8 juillet 2024
📗Check out demo notebook:https://t.co/PyfbwIlkXn
🛠️Install Yi from LlamaHub:https://t.co/MgPYfv19hI pic.twitter.com/y5EzStVsVDExciting News: we've integrated #Yi with @llama_index for more efficient retrieval and indexing capabilities, helping you build smarter and faster RAG apps with ease! Check out demo notebook: https://
docs.llamaindex.ai/en/latest/exam
ples/llm/yi/
… Install Yi from LlamaHub: https://
llamahub.ai/l/llms/llama-i
ndex-llms-yi
… -
Daily Content on Python, Data Science, ML and LLMs
By
–
That's a wrap! If you are interested in any of these topics: – Python – Data Science – Machine Learning – Data Analysis – LLMs – MLOps Find me → @Sumanth_077 I'm sharing daily content over here.
-
MLX: Deep Learning Framework for Machine Learning on macOS
By
–
5. MLX: A deeplearning framework for machine learning on macOS! You can efficiently train and deploy ML models. The repo also includes examples to run open-source models using MLX, like LLaMA, Mixtral 8x7B, LoRA, CLIP, Whisper and Stable Diffusion. https://
github.com/ml-explore/mlx -
MoE Token Independence Enables Efficient Long Context Processing in vLLM
By
–
@avshalomm solved it by utilizing the fact that actually, there is no interaction between different tokens in the MoE block, so we can iterate over the long context in chunks. This was also merged and now fixed in vLLM