23 15 99

David Mezzetti

davidmezzetti

https://github.com/davidmezzetti

AI & ML interests

None yet

Recent Activity

updated a model about 6 hours ago

NeuML/domain-labeler

liked a model about 6 hours ago

NeuML/domain-labeler

published a model about 6 hours ago

NeuML/domain-labeler

View all activity

Organizations

posted an update 3 months ago

Post

230

Do you like to "think" in Jupyter Notebooks? Want to integrate OpenCode with that? Then check this project out.

NCoder is an open-source AI coding agent that connects a local running OpenCode server to a Jupyter Notebook via TxtAI.

https://github.com/neuml/ncoder

repliedto their post 3 months ago

There is nothing inherent about it that requires a GPU. Did you run into an issue?

repliedto their post 3 months ago

Thank you!

repliedto their post 3 months ago

The training dataset is from here: https://hf.co/datasets/mecha-org/linux-command-dataset

posted an update 3 months ago

Post

1195

Tired of paying a bunch of money for large LLMs? Then read this notebook and/or watch this video on how to distill knowledge into tiny LLMs.

Notebook: https://github.com/neuml/txtai/blob/master/examples/80_Distilling_Knowledge_into_Tiny_LLMs.ipynb
Video: https://www.youtube.com/watch?v=Ol560ktgkf0

6 replies

posted an update 3 months ago

Post

2197

🥃 Distilling Tiny Embeddings. We're happy to build on the BERT Hash Series of models with this new set of fixed dimensional tiny embeddings models.

Ranging from 244K parameters to 970K and 50 dimensions to 128 dimensions these tiny models pack quite a punch.

Use cases include on-device semantic search, similarity comparisons, LLM chunking and Retrieval Augmented Generation (RAG). The advantage is that data never needs to leave the device while still having solid performance.

https://huggingface.co/blog/NeuML/bert-hash-embeddings

posted an update 4 months ago

Post

2986

🧬⚕️🔬 Encoding the World's Medical Knowledge into 970K! We're excited to release this new series of vector embeddings models for medical literature based on our recent BERT Hash work.

And you read it right, we're talking 970,000 parameters for a surprisingly strong performing model. Enjoy!

https://huggingface.co/blog/neuml/biomedbert-hash-nano

posted an update 4 months ago

Post

255

💥 Excited to publish our revamped Introducing TxtAI article using our brand new Hugging Face Teams account! 🤗

https://huggingface.co/blog/neuml/txtai

repliedto maxiw's post over 1 year ago

@maxiw Excellent work with this dataset. It inspired me to run some further analysis using graph networks and agents. See below. Thank you for putting this dataset together!

Article | Notebook | Embeddings Database

David Mezzetti

AI & ML interests

Recent Activity

Organizations

davidmezzetti's activity