This week brings significant AI news.
## OpenAI
OpenAI has unveiled its new o1 series models, `o1-preview` and `o1-mini`, which excel at complex reasoning tasks. These models allocate extra computational resources to thinking processes before generating responses.
Here are some valuable resources to learn more about these two models:
* [Introducing OpenAI o1-preview](https://openai.com/index/introducing-openai-o1-preview/)
* [Learning to Reason with LLMs](https://openai.com/index/learning-to-reason-with-llms/)
* [Let's Verify Step by Step](https://arxiv.org/abs/2305.20050)
I've also written a note with key information about this release. You can find it here: [Sep 13, Notes on OpenAI o1 series models](https://www.chengshengddeng.com/en/article/openai-o1)
## Google NoteBook LLM
Google has launched an innovative AI product called _NotebookLM_. This experimental tool harnesses the power of language models combined with existing content to generate critical insights faster. It functions as a virtual research assistant, capable of summarizing facts, explaining complex ideas, and brainstorming new connections—all based on user-selected sources. Additionally, it offers an audio feature where two AI hosts engage in a lively "deep dive" discussion about your sources at the click of a button.
To learn more about this product, check out these resources:
* [NotebookLM now lets you listen to a conversation about your sources](https://blog.google/technology/ai/notebooklm-audio-overviews/)
* [Introducing Notebook LM](https://blog.google/technology/ai/notebooklm-google-ai/)
## Anthropic
Anthropic has released their latest course on LLM prompt evaluations. These evaluations help ensure prompts are production-ready by quickly identifying edge cases and pinpointing where prompts are most effective.
You can find the course on GitHub here:
* https://github.com/anthropics/courses/tree/master/prompt_evaluations
## Mistral
Mistral released following things in the past week:
* A free tier on la Platform
* A pricing update over the entire family of models
* A new, better Mistral Small
* Free vision capabilities on le Chat with Pixtral 12B
Following is the link to the information:
* https://mistral.ai/news/september-24-release/