Code Llama: Open Foundation Models for Code https://
bit.ly/3RCyvhr A version of Llama 2, fine-tuned for generating and discussing code.
@aiatmeta
-
Code Llama: Open Foundation Models for Code Generation
By
–
-
Effective Long-Context Scaling of Foundation Models Research
By
–
Effective Long-Context Scaling of Foundation Models https://
bit.ly/3ER36jT Another piece of research that helps us build engaging conversational experiences for our AIs and the Meta AI assistant. -
Emu: Improving Image Generation with Photogenic Needles Technique
By
–
Emu: Enhancing Image Generation Models Using Photogenic Needles in a Haystack https://
bit.ly/3PVhE8r This research was an important piece of our new generative image features such as AI stickers and upcoming image editing features in Instagram like Restyle and Backdrop. -
Meta Connect 2023: AI Research and Models Behind New Features
By
–
In addition to all of the new product announcements at #MetaConnect2023, you heard about some of the research and AI models that helps bring these new features to life. Here's a look at some of the work you should know about + the research papers where you can learn more
-
Code Llama: Free Models for Research and Commercial Use
By
–
Code Llama is free for both research and commercial use and we've made three different models available:
— AI at Meta (@AIatMeta) 20 septembre 2023
– Code Llama
– Code Llama – Python
– Code Llama – Instruct
More details on each of these models and how you can download them ➡️ https://t.co/yLBEKVJhU5 pic.twitter.com/XiQRKNkFgBCode Llama is free for both research and commercial use and we've made three different models available:
– Code Llama
– Code Llama – Python
– Code Llama – Instruct More details on each of these models and how you can download them https://
bit.ly/45JiPwJ -
SeamlessM4T: Multimodal Speech Translation Model for 100 Languages
By
–
Last month we announced SeamlessM4T, a foundational multimodal model for speech translation that can perform tasks across speech-to-text, speech-to-speech and more for up to 100 languages depending on the task. More details on this work https://
bit.ly/3PqQncw -
Belebele: Multilingual Reading Comprehension Dataset for 122 Languages
By
–
We recently released Belebele, a first-of-its-kind multilingual reading comprehension dataset. It's parallel for 122 language variants, enabling direct comparison of how well models understand different languages. Dataset
-
Llama 2 and Code Llama Models Now Available on Kaggle
By
–
Llama 2 and Code Llama are now on #KaggleModels!
-
Kaggle Community Ready to Build with Llama 2
By
–
We can't wait for the Kaggle community to start building with Llama 2!
-
Excitement for SeamlessM4T Building Session
By
–
Excited for this session — we can't wait to see what people build with SeamlessM4T!