Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/huggingface/smollm
https://github.com/huggingface/smollm
Last synced: 18 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/huggingface/smollm
- Owner: huggingface
- License: apache-2.0
- Created: 2024-11-04T13:01:54.000Z (about 1 month ago)
- Default Branch: main
- Last Pushed: 2024-11-19T18:54:38.000Z (24 days ago)
- Last Synced: 2024-11-19T19:52:50.727Z (24 days ago)
- Language: Python
- Size: 41 KB
- Stars: 311
- Watchers: 9
- Forks: 15
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- StarryDivineSky - huggingface/smollm - 1.7B-Instruct，支持多种使用方式，包括 `transformers`、`trl` 和 `llama.cpp` 等工具。此外，新推出的 SmolVLM 是基于 SmolLM2 的视觉语言模型。 (A01_文本生成_文本对话 / 大语言对话模型及数据)
- awesome-vision-language-pretraining - [github
- awesome-repositories - huggingface/smollm - Everything about the SmolLM & SmolLM2 family of models (Python)
README
# SmolLM2
SmolLM2 is a family of compact language models available in three size: 135M, 360M, and 1.7B parameters. They are capable of solving a wide range of tasks while being lightweight enough to run on-device.
You can find our most capable model **🤏 SmolLM2-1.7B-Instruct** [here](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-Instruct).
**New: Introducing [SmolTalk](https://huggingface.co/datasets/HuggingFaceTB/smoltalk), the SFT dataset of SmolLM2 🚀**
## Table of Contents
1. [Usage](#usage)
- [Transformers](#transformers)
- [Chat in TRL](#chat-in-trl)
- [Local inference](#local-inference)
2. [Smol-tools](#smol-tools)
3. [Pre-training](#pre-training)
4. [Fine-tuning](#fine-tuning)
5. [Evaluation](#evaluation)
6. [Synthetic data pipelines](#synthetic-data-pipelines)
## Usage
Our most powerful model is `SmolLM2-1.7B-Instruct`, which you can use as an assistant with `transformers`, `trl`, or using quantized versions with tools like `llama.cpp`, `MLX`, and `transformers.js`. For lighter applications, you can also use the smaller models `SmolLM2-360M` and`SmolLM2-135M`, which are suitable for on-device usage and can be integrated similarly.
All available in this [collection](https://huggingface.co/collections/HuggingFaceTB/smollm2-6723884218bcda64b34d7db9).
### Transformers
```bash
pip install transformers
```
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
checkpoint = "HuggingFaceTB/SmolLM2-1.7B-Instruct"
device = "cuda" # for GPU usage or "cpu" for CPU usage
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
# for multiple GPUs install accelerate and do `model = AutoModelForCausalLM.from_pretrained(checkpoint, device_map="auto")`
model = AutoModelForCausalLM.from_pretrained(checkpoint).to(device)
messages = [{"role": "user", "content": "Write a 100-word article on 'Benefits of Open-Source in AI research"}]
input_text=tokenizer.apply_chat_template(messages, tokenize=False)
inputs = tokenizer.encode(input_text, return_tensors="pt").to(device)
outputs = model.generate(inputs, max_new_tokens=50, temperature=0.2, top_p=0.9, do_sample=True)
print(tokenizer.decode(outputs[0]))
```
### Chat in TRL
You can also use the TRL CLI to chat with the model from the terminal:
```bash
pip install trl
trl chat --model_name_or_path HuggingFaceTB/SmolLM2-1.7B-Instruct --device cpu
```
You can find more details on how to leverage the model for use cases such as text summarization, text rewriting and function calling in the model card: https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-Instruct
### Local inference
You can use the models locally with frameworks like `llama.cpp`, `MLX`, and `transformers.js`. You can find the instructions to run SmolLM2 with these frameworks at [local-inference](local_inference/README.md).
All models are available in this [collection](https://huggingface.co/collections/HuggingFaceTB/smollm2-6723884218bcda64b34d7db9).
## Smol-tools
A collection of lightweight AI-powered tools built with LLaMA.cpp and small language models. These tools are designed to run locally on your machine without requiring expensive GPU resources.
Further instructions on how to use the tools can be found in the [smol-tools README](smol_tools/README.md).
## Pre-training
You can find scripts for launching pre-training with [nanotron](https://github.com/huggingface/nanotron/) under [pre-training](pre-training/README.md), we share the exact configs for training SmolLM1 and will upload SmolLM2's configs soon.
## Fine-tuning
You can find an example script to finetune SmolLM2 using `TRL` and `PEFT` in the `finetuning` folder. We also link to our post-training scripts for SmolLM2 using the alignment handbook.
## Evaluation
You can find more detailed evaluation of each model size in the model cards in this [collection](https://huggingface.co/collections/HuggingFaceTB/smollm2-6723884218bcda64b34d7db9).
We use [lighteval](https://github.com/huggingface/lighteval) for all our evaluations, for more details refer to the [evaluation README](evaluation/README.md).
## Synthetic data pipelines
We released [SmolTalk](https://huggingface.co/datasets/HuggingFaceTB/smoltalk) the SFT dataset used for building SmolLM2 instruct models. It was created with [distilabel](https://github.com/argilla-io/distilabel) and you can check and execute the synthetic data pipelines in [distilabel_pipelines README](distilabel_pipelines/README.md)
Comparison of models finetuned on SmolTalk and Orca AgentInstruct 1M. For more details, refer to the dataset card.