KissanAI Releases Dhenu Llama 3, an Indic LLM for Farmers

KissanAI has released Dhenu Llama 3, fine-tuned on Llama3 8B, as announced by KissanAI founder Pratik Desai.

"It is available for anyone to tinker with and provide feedback. Feel free to host and share if you have a spare GPU. We will have an instruction version with a dataset five times larger in the near future," wrote Desai on LinkedIn.

Meta recently released Llama 3. The model is available in 8B and 70B parameter versions and has been trained on over 15 trillion tokens, making it seven times larger than Llama 2’s dataset. Llama 3 provides enhanced reasoning and coding capabilities, and its training process is three times more efficient than its predecessor.

The models are now also available on Hugging Face.

Interestingly, In its Community License Agreement conditions, the company has mentioned, “If you use the Llama Materials to create, train, fine-tune, or otherwise improve an AI model, which is distributed or made available, you shall also include “Llama 3” at the beginning of any such AI model name”, under the redistribution and use section.

Meta is also training a model with more than 400 billion parameters which Mark Zuckerberg said in a Reel on Instagram is going to be the top performing model out there.

The 7B models outperforms Gemma and Mistral on all benchmarks and the 70B model outperforms Gemini Pro 1.5 and Claude 3 Sonnet.

Llama 3 models are now rolling out on Amazon SageMaker, Databricks, Google Cloud, Hugging Face, Kaggle, IBM WatsonX, Microsoft Azure, NVIDIA NIM, and Snowflake. Additionally, the models will be compatible with hardware platforms provided by AMD, AWS, Dell, Intel, NVIDIA, and Qualcomm.

Last year, KissanAI released Dhenu 1.0, a Agriculture LLM for Indian farming built on Llama 2. This model understands English, Hindi, and Hinglish, catering to farmers' language preferences. It is trained on extensive datasets and processes 300,000 instructions in both languages, enhancing support for farmers' queries.