Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
AI Model Name: Llama 3 70B "Built with Meta Llama 3" https://llama.meta.com/llama3/license/
|
| 2 |
+
|
| 3 |
+
This is the result of running AutoAWQ to quantize the LLaMA-3 70B model to ~4 bits/parameter.
|
| 4 |
+
|
| 5 |
+
To launch an OpenAI-compatible API endpoint on your Linux server with 2x 3090 or 4090 GPUs:
|
| 6 |
+
|
| 7 |
+
```
|
| 8 |
+
git lfs install
|
| 9 |
+
git clone https://huggingface.co/catid/cat-llama-3-70b-awq-q256-w4-gemm
|
| 10 |
+
|
| 11 |
+
conda create -n vllm70 python=3.10 -y && conda activate vllm70
|
| 12 |
+
|
| 13 |
+
pip install -U git+https://github.com/vllm-project/vllm.git
|
| 14 |
+
|
| 15 |
+
python -m vllm.entrypoints.openai.api_server --model cat-llama-3-70b-awq-q256-w4-gemm --tensor-parallel-size 2 --gpu-memory-utilization 0.935
|
| 16 |
+
```
|
| 17 |
+
|
| 18 |
+
Sadly this *barely* doesn't fit by ~300MB or so.
|