22 91 92

Andrew Reed

andrewrreed

https://www.andrewreed.com

AI & ML interests

Applied ML, Practical AI, Inference & Deployment, LLMs, Multi-modal Models, RAG

Recent Activity

upvoted a collection 5 days ago

Kimi-VL-A3B

liked a Space 5 days ago

xet-team/repo-graph

reacted to jsulz's post with 🔥 8 days ago

Huge week for https://huggingface.co/xet-team as Llama 4 is the first major model on Hugging Face uploaded with Xet providing the backing! Every byte downloaded comes through our infrastructure. Using Xet on Hugging Face is the fastest way to download and iterate on open source models and we've proved it with Llama 4 giving a boost of ~25% across all models. We expect builders on the Hub to see even more improvements, helping power innovation across the community. With the models on our infrastructure, we can peer in and see how well our dedupe performs across the Llama 4 family. On average, we're seeing ~25% dedupe, providing huge savings to the community who iterate on these state-of-the-art models. The attached image shows a few selected models and how they perform on Xet. Thanks to the https://huggingface.co/meta-llama team for launching on Xet!

View all activity

Organizations

andrewrreed's activity

upvoted a collection 5 days ago

Kimi-VL-A3B

Collection

Moonshot's efficient MoE VLMs, exceptional on agent, long-context, and thinking • 6 items • Updated 2 days ago • 58

liked a Space 5 days ago

Repo Graph

👁

A byte-level map of the Hugging Face Hub

reacted to jsulz's post with 🔥 8 days ago

Post

3562

Huge week for

xet-team as Llama 4 is the first major model on Hugging Face uploaded with Xet providing the backing! Every byte downloaded comes through our infrastructure.

Using Xet on Hugging Face is the fastest way to download and iterate on open source models and we've proved it with Llama 4 giving a boost of ~25% across all models.

We expect builders on the Hub to see even more improvements, helping power innovation across the community.

With the models on our infrastructure, we can peer in and see how well our dedupe performs across the Llama 4 family. On average, we're seeing ~25% dedupe, providing huge savings to the community who iterate on these state-of-the-art models. The attached image shows a few selected models and how they perform on Xet.

Thanks to the

meta-llama team for launching on Xet!

upvoted a collection 9 days ago

Llama 4

Collection

Llama 4 release • 10 items • Updated 9 days ago • 423

liked a model 9 days ago

meta-llama/Llama-4-Scout-17B-16E-Instruct

Image-Text-to-Text • Updated 5 days ago • 541k • • 764

upvoted a paper 11 days ago

YourBench: Easy Custom Evaluation Sets for Everyone

Paper • 2504.01833 • Published 12 days ago • 16

upvoted an article 21 days ago

Article

The New and Fresh analytics in Inference Endpoints

24 days ago

• 19

updated a collection 25 days ago

Awesome Spaces

Collection

28 items • Updated 25 days ago • 4

liked a Space 25 days ago

411

Gemini Co-Drawing

✏

Gemini 2.0 native image generation co-doodling

updated a collection 27 days ago

Awesome Spaces

Collection

28 items • Updated 25 days ago • 4

liked a Space 27 days ago

398

Stable Virtual Camera

⚡

Generate virtual camera views from input images

upvoted an article 27 days ago

Article

Xet is on the Hub

27 days ago

• 46

reacted to MJannik's post with 🔥 about 1 month ago

Post

1752

I've published an article showing five ways to use 🪢 Langfuse with 🤗 Hugging Face.

My personal favorite is Method #4: Using Hugging Face Datasets for Langfuse Dataset Experiments. This lets you benchmark your LLM app or AI agent with a dataset hosted on Hugging Face. In this example, I chose the GSM8K dataset ( openai/gsm8k) to test the mathematical reasoning capabilities of my smolagent :)

Link to the Article here on HF: https://huggingface.co/blog/MJannik/hugging-face-and-langfuse