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.
| Date | Stars |
|---|---|
| 2026-07-31 | 711 |
| 2026-08-03 | 711 |
| 2026-08-04 | 711 |
| 2026-08-06 | 711 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Hugging Face Llama Recipes
🤗🦙Welcome! This repository contains *minimal* recipes to get started quickly
with **Llama 3.x** models, including **Llama 3.1**, **Llama 3.2**, and **Llama 3.3**.
* To get an overview of Llama 3.1, please visit the [Hugging Face announcement blog post (3.1)](https://huggingface.co/blog/llama31).
* To get an overview of Llama 3.2, please visit the [Hugging Face announcement blog post (3.2)](https://huggingface.co/blog/llama32).
* For more advanced end-to-end use cases with open ML, please visit the [Open Source AI Cookbook](https://huggingface.co/learn/cookbook/index).
This repository is WIP so that you might see considerable changes in the coming days.
> [!NOTE]
> To use Llama 3.x, you need to accept the license and request permission
to access the models. Please visit [the Hugging Face repos](https://huggingface.co/meta-llama)
and submit your request. You only need to do this once per collection; you'll get access to
all the repos in the collection if your request is approved.
## Getting Started
The easiest way to quickly run a Llama 🦙 on your machine would be with the
🤗 `transformers` repository. Make sure you have the latest release installed.
```shell
$ pip install -U transformers
```
Let us conversate with an instruction tuned model.
```python
import torch
from transformers import pipeline
device = "cuda" if torch.cuda.is_available() else "cpu"
llama_31 = "meta-llama/Llama-3.1-8B-Instruct" # <-- llama 3.1
llama_32 = "meta-llama/Llama-3.2-3B-Instruct" # <-- llama 3.2
prompt = [
{"role": "system", "content": "You are a helpful assistant, that responds as a pirate."},
{"role": "user", "content": "What's Deep Learning?"},
]
generator = pipeline(model=llama_32, device=device, torch_dtype=torch.bfloat16)
generation = generator(
prompt,
do_sample=False,
temperature=1.0,
top_p=1,
max_new_tokens=50
)
print(f"Generation: {generation[0]['generated_text']}")
# Generation:
# [
# {'role': 'system', 'content': 'You are a helpful assistant, that responds as a pirate.'},
# {'role': 'user', 'content': "What's Deep Learning?"},
# {'role': 'assistant', 'content': "Yer lookin' fer a treasure trove o'
# knowledge on Deep Learnin', eh? Alright then, listen close and
# I'll tell ye about it.\n\nDeep Learnin' be a type o' machine
# learnin' that uses neural networks"}
# ]
```
## Local Inference
Would you like to run inference of the Llama models locally?
So do we! The memory requirements depend on the model size and the
precision of the weights. Here's a table showing the approximate
memory needed for different configurations:
| Model Size | Llama Variant | BF16/FP16 | FP8 | INT4(AWQ/GPTQ/bnb) |
| :--: | :--: | :--: | :--: | :--: |
| 1B | 3.2 | 2.5 GB | 1.25GB | 0.75GB |
| 3B | 3.2 |6.5 GB | 3.2GB | 1.75GB |
| 8B | 3.1 |16 GB | 8GB | 4GB |
| 70B | 3.1 and 3.3 | 140 GB | 70GB | 35GB |
|405B | 3.1 |810 GB | 405GB | 204GB |
> [!NOTE]
> These are estimated values and may vary based on specific
implementation details and optimizations.
Working with the capable Llama 3.1 8B models:
* [Run Llama 3.1 8B in 4-bits with bitsandbytes](../local_inference/4bit_bnb.ipynb)
* [Run Llama 3.1 8B in 8-bits with bitsandbytes](../local_inference/8bit_bnb.ipynb)
* [Run Llama 3.1 8B with AWQ & fused ops](../local_inference/awq.ipynb)
Working on the 🐘 big Llama 3.1 405B model:
* [Run Llama 3.1 405B FP8](../local_inference/fp8-405B.ipynb)
* [Run Llama 3.1 405B quantized to INT4 with AWQ](../local_inference/awq_generation.py)
* [Run Llama 3.1 405B quantized to INT4 with GPTQ](../local_inference/gptq_generation.py)
## Model Fine Tuning:
It is often not enough to run inference on the model.
Many times, you need to fine-tune the model on some
custom dataset. Here are some scripts showing
how to fine-tune the models.
Fine tune models on your custom dataset:
* [Fine tune Llama 3.2 VisiExcerpt of 7,868 characters
Read on GitHub
Marc Sun · Hugging Face
15
Sergio Paniego Blanco · @huggingface · Spain
15
Lysandre Debut · Hugging Face · France
14
vb · @openai
13
Arthur · @huggingface · France
12
Omar Sanseviero · Google · Switzerland
10
Alvaro Bartolome · @huggingface · Spain
8
Pedro Cuenca · Spain
8
Merve Noyan · @huggingface · France
7
Joshua Lochner · @huggingface
7
Aritra Roy Gosthipaty · @huggingface · India
5
Sakalya Mitra · TheAgentic · India
5
3
elie · @PrimeIntellect-ai · France
3
@huggingface · Switzerland
2
1
Gabriel Martín Blázquez · @supersonik-ai · Spain
1
célina · @huggingface
1
1
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:dfab16f2a9376f7c, llm:Repository name 'huggingface-llama-recipes' and owner Hugging Face imply collection of fine-tuning/training recipes for LLaMA models (practical notebooks/scripts for model training and fine-tuning). Language notebooks.
matched fp:dfab16f2a9376f7c, llm:Repository name 'huggingface-llama-recipes' and owner Hugging Face imply collection of fine-tuning/training recipes for LLaMA models (practical notebooks/scripts for model training and fine-tuning). Language notebooks.