Instructions to use SpectraSuite/QuantLM_830M_4bit_Unpacked with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SpectraSuite/QuantLM_830M_4bit_Unpacked with Transformers:

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

pipe = pipeline("text-generation", model="SpectraSuite/QuantLM_830M_4bit_Unpacked")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SpectraSuite/QuantLM_830M_4bit_Unpacked")
model = AutoModelForCausalLM.from_pretrained("SpectraSuite/QuantLM_830M_4bit_Unpacked")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use SpectraSuite/QuantLM_830M_4bit_Unpacked with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SpectraSuite/QuantLM_830M_4bit_Unpacked"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SpectraSuite/QuantLM_830M_4bit_Unpacked",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/SpectraSuite/QuantLM_830M_4bit_Unpacked

SGLang

How to use SpectraSuite/QuantLM_830M_4bit_Unpacked 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 "SpectraSuite/QuantLM_830M_4bit_Unpacked" \
    --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": "SpectraSuite/QuantLM_830M_4bit_Unpacked",
		"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 "SpectraSuite/QuantLM_830M_4bit_Unpacked" \
        --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": "SpectraSuite/QuantLM_830M_4bit_Unpacked",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use SpectraSuite/QuantLM_830M_4bit_Unpacked with Docker Model Runner:
```
docker model run hf.co/SpectraSuite/QuantLM_830M_4bit_Unpacked
```

QuantLM_830M_4bit_Unpacked / README.md

tejasexpress

Update README.md

b0a4efe verified almost 2 years ago

preview code

raw

history blame contribute delete

854 Bytes

metadata

license: apache-2.0

QuantLM 830M 4 bit

QuantLM, unpacked to FP16 format - compatible with FP16 GEMMs. After unpacking, QuantLM has the same architecture as LLaMa.

import transformers as tf, torch
model_name = "SpectraSuite/QuantLM_830M_4bit_Unpacked"
# Please adjust the temperature, repetition penalty, top_k, top_p and other sampling parameters according to your needs.
pipeline = tf.pipeline("text-generation", model=model_id, model_kwargs={"torch_dtype": torch.float16}, device_map="auto")
# These are base (pretrained) LLMs that are not instruction and chat tuned. You may need to adjust your prompt accordingly.
pipeline("Once upon a time")

License: Apache 2.0
We will use our GitHub repo for communication (including HF repo related queries). Feel free to open an issue here https://github.com/NolanoOrg/SpectraSuite