Super proud of the team behind Llama 3.1 8B storm model and it’s now available on @monsterapis for finetuning and deployment in a single click!
Why it matters?
This model is beating all the benchmarks through the roof! With their self curation technique, it is helping the model perform better than other source LLMs and reaching closer to GPT 4o with just a 8 Billion parameter model is not a small feat and now you can customise and improve it further for your task like summarization or chatbot tuning on MonsterAPI.
Let’s go 🤘🚀
#llm
⭐ Today marks 1-month since the release of Llama-3.1-Storm-8B model with 52k+ collective downloads with model availability on @huggingface 🤗, @ollama , and @UnslothAI . We are starting a discord group for Storm: https://t.co/jATbwGWTC5
🚀 From today onwards, model is also accessible via @monsterapis for 1-click fine-tuning and deployment:
- Finetune Llama-3.1-Storm-8B model: https://t.co/WBkAOSfUyP
- 1-click deployment of Llama-3.1-Storm-8B and your fine-tuned Storm model as API endpoint on Monster API: https://t.co/BQwkAhrWv8
Monster API team has put together a easy-to-follow Colab notebook for fine-tuning Llama-3.1-Storm-8B model on your dataset: https://t.co/Bb53zTkaZR
This is our first generalist model and we are grateful to developers and companies who are playing with Storm for their use-cases! 😄 🙏
@sunglassesface@mattshumer_ Self-Curation filters data before fine-tuning but it's normal fine-tuning. While Self-Reflection is one type of Chain-Of-Thought fine-tuning where model generate structured rationale before generating output and its said that fine-tuning is done that way.