Fuck it – it’s raining smol LMs – SmolLM2 1.7B – beats Qwen 2.5 1.5B & Llama 3.21B, Apache 2.0 licensed, trained on 11 Trillion tokens > 135M, 360M, 1.7B parameter model
> Trained on FineWeb-Edu, DCLM, The Stack, along w/ new mathematics and coding datasets
> Specialises in
LLMS
-
SmolLM2 1.7B Beats Larger Models with Apache 2.0 License
By
–
-
LLM Agentic AI Workflows Presented at Turkey AI Summit
By
–
Our team was honored to participate in the Turkiye AI Summit in Istanbul, where Alex White presented a session on how to make use of #LLM Agentic #AI workflows. It was a pleasure to speak with industry leaders and share our insights.
-
ChatGPT Search App Built with Groq and Gradio
By
–
chatgpt search built a app with groq-gradio which was just released today @GroqInc @Gradio
-
OpenAI Uses o1 Synthetic Data for Model Training
By
–
First time OpenAI officially says they are using synthetic data from o1 to train their models. SearchGPT was trainier by o1. Makes you think how GPT-5 was trained 😉
-
ChatGPT Search Global Release: Strong Competition Against Perplexity
By
–
ChatGPT search: global release. It’s already available for me and holy moly is it good! That’s the direct attack on Perplexity and Google! Gonna test it further but so far: cudos Openai! pic.twitter.com/QwahW3fwDe
— Chubby♨️ (@kimmonismus) 31 octobre 2024ChatGPT search: global release. It’s already available for me and holy moly is it good! That’s the direct attack on Perplexity and Google! Gonna test it further but so far: cudos Openai!
-
Turbo LoRA Enhancement: 2x Throughput Boost Without Retraining
By
–
-
AI Prompt Engineering: A Deep Dive into LLM Optimization
By
–
AI prompt engineering: A deep dive https://
bit.ly/3ZNUExp
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
Highlights from OpenAI DevDay: image model, o1 coding paradigm, Realtime API
By
–
> Highlights from yesterday’s @OpenAI DevDay in London: image model in the works, o1’s new coding paradigm, Realtime API o1 creates a new paradigm for code. OpenAI devs seem to have massively adopted o1 + Cursor, seems like the experience is vastly improved compared to GH
