Instructions to use Skywork/Skywork-13B-Math-8bits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Skywork/Skywork-13B-Math-8bits with Transformers:

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

pipe = pipeline("text-generation", model="Skywork/Skywork-13B-Math-8bits", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Skywork/Skywork-13B-Math-8bits", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Skywork/Skywork-13B-Math-8bits with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Skywork/Skywork-13B-Math-8bits"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Skywork/Skywork-13B-Math-8bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Skywork/Skywork-13B-Math-8bits

SGLang

How to use Skywork/Skywork-13B-Math-8bits 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 "Skywork/Skywork-13B-Math-8bits" \
    --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": "Skywork/Skywork-13B-Math-8bits",
		"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 "Skywork/Skywork-13B-Math-8bits" \
        --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": "Skywork/Skywork-13B-Math-8bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Skywork/Skywork-13B-Math-8bits with Docker Model Runner:
```
docker model run hf.co/Skywork/Skywork-13B-Math-8bits
```

zhao1iang commited on Oct 30, 2023

Commit

abb46d0

1 Parent(s): b351f9e

Update README.md

Browse files

Files changed (1) hide show

README.md +6 -0

README.md CHANGED Viewed

@@ -43,6 +43,12 @@ license_link: >-
 **Skywork-13B-Math**: Skywork-13B-Math model has undergone specialized training to enhance its mathematical abilities. In the 13B-scale model, the Skywork-13B-Math model ranked first in the GSM8K evaluation, and it also performed exceptionally well on the MATH dataset and CMATH, placing it among the top-level 13B models.
 如果您希望了解更多的信息，如训练方案，评估方法，请参考我们的[技术报告](https://arxiv.org/skywork-tech-report)，[Skymath](https://arxiv.org/abs/2310.16713)论文，[SkyworkMM](https://github.com/will-singularity/Skywork-MM/blob/main/skywork_mm.pdf)论文。
 If you are interested in more training and evaluation details, please refer to our [technical report](https://arxiv.org/skywork-tech-report), [Skymath]((https://arxiv.org/skywork-tech-report)) paper and [SkyworkMM](https://github.com/will-singularity/Skywork-MM/blob/main/skywork_mm.pdf) paper.

 **Skywork-13B-Math**: Skywork-13B-Math model has undergone specialized training to enhance its mathematical abilities. In the 13B-scale model, the Skywork-13B-Math model ranked first in the GSM8K evaluation, and it also performed exceptionally well on the MATH dataset and CMATH, placing it among the top-level 13B models.
+**Skywork-13B-Math-8bits**模型为**Skywork-13B-Math**的8bits量化版，支持用户在消费级显卡上进行进行部署和推理。
+**Skywork-13B-Math-8bits** is a quantizated model of **Skywork-13B-Math**, to support deployment and inference on consumer-grade GPUs.
 如果您希望了解更多的信息，如训练方案，评估方法，请参考我们的[技术报告](https://arxiv.org/skywork-tech-report)，[Skymath](https://arxiv.org/abs/2310.16713)论文，[SkyworkMM](https://github.com/will-singularity/Skywork-MM/blob/main/skywork_mm.pdf)论文。
 If you are interested in more training and evaluation details, please refer to our [technical report](https://arxiv.org/skywork-tech-report), [Skymath]((https://arxiv.org/skywork-tech-report)) paper and [SkyworkMM](https://github.com/will-singularity/Skywork-MM/blob/main/skywork_mm.pdf) paper.