Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
A New Tamil Large Language Model (LLM) Based on Llama 2
| Date | Stars |
|---|---|
| 2026-07-31 | 328 |
| 2026-08-06 | 329 |
Today
+1 stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Tamil-Llama: A Family of LLaMA-based LLMs focused on Tamil Language <img src="assets/introducing_tamil_llama.png" alt="Tamil LLaMA Image" width="300" height="auto"> ## Description This repository contains the code and models for "Tamil-Llama", a project focused on enhancing the performance of language models for the Tamil language. It builds upon the open-source LLaMA model, introducing additional Tamil tokens and employing the LoRA methodology for efficient training. Please read the technical report for more details. Technical Report: [https://arxiv.org/abs/2311.05845](https://arxiv.org/abs/2311.05845) If you appreciate this work and would like to support its continued development, consider [buying me a coffee](https://www.buymeacoffee.com/abhinand.b). Your support is invaluable and greatly appreciated. [](https://www.buymeacoffee.com/abhinand.b) ## Updates ### Feb 25, 2024 Google's Gemma 2B Model was adapter for Tamil (Experimental Release) based on the same framework with a few changes. More info in [this](https://www.linkedin.com/posts/abhinand-05_%3F%3F%3F%3F%3F%3F%3F%3F%3F%3F%3F-%3F%3F%3F%3F%3F-%3F%3F-activity-7167767094619430912-VspR?utm_source=share&utm_medium=member_desktop) LinkedIn post. > **Note:** I have migrated to [Llama-Factory](https://github.com/hiyouga/LLaMA-Factory) for pretraining and [Axolotl](https://github.com/OpenAccess-AI-Collective/axolotl) for finetuning. - No expansion in vocab for Gemma as it already has 256k vocab size and minnescule amounts of Tamil tokens. - Continually pretrain on all available Tamil Wikipedia data for 3 epochs. - Finetune on Tamil Alpaca + English Alpaca mix for 5 epochs - Model tops Open LLM Leaderboard for models under 3B params as of Feb 2023. **Download Links:** - [Tamil Gemma 2B Alpha](https://huggingface.co/abhinand/gemma-2b-it-tamil-v0.1-alpha) - [Tamil Gemma 2B Alpha GGUF](https://huggingface.co/abhinand/gemma-2b-it-tamil-v0.1-alpha-GGUF) ### Jan 23, 2024 For more details, please read the detailed blog post [here](https://abhinand05.medium.com/breaking-language-barriers-introducing-tamil-llama-v0-2-and-its-expansion-to-telugu-and-malayalam-deb5d23e9264). - Tamil LLaMA v0.2 models are out. It is a significant upgrade compared to the earlier version. - Tamil LLaMA is now bilingual, it can fluently respond in both English and Tamil. - Better tokenizer. - Better base model. - Better fine tuning dataset and performance. - Our models match or betters the performance of Meta's LLaMA 2 is almost all the benchmarks. - Following the same methodology the first ever Telugu and Malayam LLaMA models are also released. ## Table of Contents - [Available Models](#available-models) - [Benchmark Scores](#benchmark-scores) - [Demo](#demo) - [Getting Started](#getting-started) - [Datasets](#datasets) - [Prompting Format](#prompting-format-for-instruction-models) - [Usage Note](#usage-note) - [Contributions](#contributions) - [License](#license) - [Citation](#citation) - [Contact](#contact) ## Available Models | Model | Type | Data | Base Model | # Params | Download Links | |--------------------------|-----------------------------|-------------------|----------------------|------|------------------------------------------------------------------------| | Tamil LLaMA 7B Base | Base model | 12GB | LLaMA 7B | 7B | [HF Hub](https://huggingface.co/abhinand/tamil-llama-7b-base-v0.1) | | Tamil LLaMA 13B Base | Base model | 4GB | LLaMA 13B | 13B | [HF Hub](https://huggingface.co/abhinand/tamil-llama-13b-base-v0.1) | | Tamil LLaMA 7B Instruct | Instruction following model | 145k instructions | Tamil LLaMA 7B Base | 7B | [HF Hub](https://huggingfac
Excerpt of 12,084 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:87661c2ab5a1cf8e, llm:description: 'A New Tamil Large Language Model (LLM) Based on Llama 2'
matched fp:87661c2ab5a1cf8e, llm:description: 'A New Tamil Large Language Model (LLM) Based on Llama 2'
matched fp:87661c2ab5a1cf8e, llm:description: 'A New Tamil Large Language Model (LLM) Based on Llama 2'