๐จ Introducing Gaja ๐จ ~ (Llama3-Gaja)
A series of open source bilingual Hindi-English LLMs finetuned on top of Llama3-8b by @AIatMeta
huggingface.co/Cognitive-Lab/โฆ
CognitiveLab
27 posts
CognitiveLab
@cognitivelab_ai
Democratizing Generative AI
- ๐จNew Indic LLM alert๐จ Introducing Ambari โ A series of Open Source Bilingual Kannada-English Large Language Models! Ambari tackles the challenge of adapting LLMs for indic languages, starting with Kannada.
- ๐ Thrilled to announce CognitiveLab wins the six-figure Llama Impact Grant by @Meta ! ๐ฎ๐ณ As India's only recipient, we're powering Nayana a multimodal, multilingual multi task AI model family! #AI #MultilingualAI #Innovation
- Introducing ๐จIndic LLM Leaderboard๐จ (alpha release)
- Replying to @cognitivelab_ai
- Replying to @cognitivelab_aiHere is the blog going into details about the model cognitivelab.in/blog/introduciโฆ You can find the models here huggingface.co/collections/Coโฆ
- Replying to @cognitivelab_aiMeet Nayana ("eyes" in Sanskrit): A unified AI model for text, vision, & audio! โ Supports 22 languages (10 Indic + 12 global) โ OCR, translation, Q&A, summarization many more tasks Know more :
- Replying to @cognitivelab_aiNayanaโs Mission: Democratize AI for education, healthcare, governance, & cultural preservation across languages & modalities. Official Announcement :
- Replying to @cognitivelab_aiOur inaugural models, ๐๐บ๐ฏ๐ฎ๐ฟ๐ถ-๐ณ๐-๐ฏ๐ฎ๐๐ฒ-๐๐ฌ.๐ญ and ๐๐บ๐ฏ๐ฎ๐ฟ๐ถ-๐ณ๐-๐๐ป๐๐๐ฟ๐๐ฐ๐-๐๐ฌ.๐ญ, achieve impressive results on a compact 1 billion-token training dataset, trained across multiple stages.
- Replying to @cognitivelab_ai@cognitivelab_ai latest Milestones: ๐ Poster at #llamacon ๐ Papers accepted at NAACL & CVPR workshops ๐ Training on millions of synthetic datasets We're replacing fragmented AI pipelines with ONE cohesive model!
- Replying to @cognitivelab_ai๐๐ผ Finetune for multi turn chat conversation ๐๐ผ Performers decent in cross lingual tasks ๐๐ผ Will soon put out the evals on Indic LLM leader board
- Replying to @cognitivelab_aiFeatures include: ๐๐ผ Support for 7 Indic languages. ๐๐ผ Open source, hosted on Hugging Face. ๐๐ผ Support for 4 Indic benchmarks, with more to be added. ๐๐ผ Seamless integration with indic_eval This is the alpha release and we will be adding a lot more tested features soon
- Replying to @kn_neeraj @paraschopra and 2 othersMost of our work is open source, and we'll keep updating our website with the latest updates. We are active on GitHub and HuggingFace too. cognitivelab.in
- Replying to @cognitivelab_aiOur first attempt at fine-tuning LLama3 for Indic languages. we choose Hindi as LLama3 was efficient at tokenising it Many more experiments/iterations are in progress
