Text Generation
Transformers
Safetensors
English
llama
tiny-model
sub-1M
cpu
small
tiny
quark
1m
text-generation-inference
Instructions to use LH-Tech-AI/Quark-0.5M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use LH-Tech-AI/Quark-0.5M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="LH-Tech-AI/Quark-0.5M")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("LH-Tech-AI/Quark-0.5M") model = AutoModelForCausalLM.from_pretrained("LH-Tech-AI/Quark-0.5M") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use LH-Tech-AI/Quark-0.5M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "LH-Tech-AI/Quark-0.5M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "LH-Tech-AI/Quark-0.5M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/LH-Tech-AI/Quark-0.5M
- SGLang
How to use LH-Tech-AI/Quark-0.5M with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "LH-Tech-AI/Quark-0.5M" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "LH-Tech-AI/Quark-0.5M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "LH-Tech-AI/Quark-0.5M" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "LH-Tech-AI/Quark-0.5M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use LH-Tech-AI/Quark-0.5M with Docker Model Runner:
docker model run hf.co/LH-Tech-AI/Quark-0.5M
Update README.md
Browse files
README.md
CHANGED
|
@@ -17,7 +17,7 @@ tags:
|
|
| 17 |
- 1m
|
| 18 |
---
|
| 19 |
|
| 20 |
-
# Quark-0.5M
|
| 21 |
|
| 22 |
**Quark-0.5M** is an ultra-lightweight Llama-based model with only **465,504 parameters**.
|
| 23 |
It was trained from scratch to demonstrate the power of high-quality data (FineWeb-Edu) on extremely small architectures.
|
|
@@ -39,7 +39,7 @@ It was trained from scratch to demonstrate the power of high-quality data (FineW
|
|
| 39 |
- **Optimizer:** AdamW with Cosine Learning Rate Decay
|
| 40 |
|
| 41 |
## Intended Use
|
| 42 |
-
Quark is a research project to explore the limits of "Micro-LLMs". It is surprisingly capable of forming grammatically correct English sentences and structured lists, despite fitting into less than
|
| 43 |
|
| 44 |
## Performance Example
|
| 45 |
> **Prompt:** "Artificial intelligence is "
|
|
|
|
| 17 |
- 1m
|
| 18 |
---
|
| 19 |
|
| 20 |
+
# 🔬 Quark-0.5M
|
| 21 |
|
| 22 |
**Quark-0.5M** is an ultra-lightweight Llama-based model with only **465,504 parameters**.
|
| 23 |
It was trained from scratch to demonstrate the power of high-quality data (FineWeb-Edu) on extremely small architectures.
|
|
|
|
| 39 |
- **Optimizer:** AdamW with Cosine Learning Rate Decay
|
| 40 |
|
| 41 |
## Intended Use
|
| 42 |
+
Quark is a research project to explore the limits of "Micro-LLMs". It is surprisingly capable of forming grammatically correct English sentences and structured lists, despite fitting into less than 2MB of disk space.
|
| 43 |
|
| 44 |
## Performance Example
|
| 45 |
> **Prompt:** "Artificial intelligence is "
|