Instructions to use dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq

SGLang

How to use dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq 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 "dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq" \
    --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": "dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq",
		"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 "dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq" \
        --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": "dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq with Docker Model Runner:
```
docker model run hf.co/dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq
```

mobicham commited on Feb 5, 2025

Commit

e4b888a

verified ·

1 Parent(s): 195ffb9

Update README.md

Browse files

Files changed (1) hide show

README.md +9 -4

README.md CHANGED Viewed

@@ -58,17 +58,22 @@ from hqq.core.quantize import *
 from hqq.utils.patching import *
 from hqq.utils.generation_hf import HFGenerator
-backend = 'bitblas' #"bitblas" #None
 #Load the model
 ###################################################
 model_id = 'mobiuslabsgmbh/Llama-3-8b-instruct_2bitgs64_hqq'
-model     = AutoHQQHFModel.from_quantized(model_id, cache_dir='.', compute_dtype=torch.float16, adapter='adapter_v0.1.lora').eval();
-tokenizer = AutoTokenizer.from_pretrained(model_id)
 #Use optimized inference kernels
 ###################################################
-prepare_for_inference(model, backend=backend, allow_merge=False) #It takes a while...
 #Generate
 ###################################################

 from hqq.utils.patching import *
 from hqq.utils.generation_hf import HFGenerator
+#Settings
+###################################################
+backend       = "bitblas" #bitblas or gemlite for 2-bit runtime
+compute_dtype = torch.bfloat16 if backend=="torchao_int4" else torch.float16
+device        = 'cuda:0'
+cache_dir     = '.'
 #Load the model
 ###################################################
 model_id = 'mobiuslabsgmbh/Llama-3-8b-instruct_2bitgs64_hqq'
+model     = AutoHQQHFModel.from_quantized(model_id, cache_dir=cache_dir, compute_dtype=compute_dtype, device=device, adapter='adapter_v0.1.lora').eval();
+tokenizer = AutoTokenizer.from_pretrained(model_id, cache_dir=cache_dir)
 #Use optimized inference kernels
 ###################################################
+prepare_for_inference(model, backend=backend) #It takes a while...
 #Generate
 ###################################################