Instructions to use AtAndDev/ShortKing-3b-v0.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AtAndDev/ShortKing-3b-v0.2 with Transformers:

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

pipe = pipeline("text-generation", model="AtAndDev/ShortKing-3b-v0.2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AtAndDev/ShortKing-3b-v0.2")
model = AutoModelForCausalLM.from_pretrained("AtAndDev/ShortKing-3b-v0.2")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use AtAndDev/ShortKing-3b-v0.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AtAndDev/ShortKing-3b-v0.2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AtAndDev/ShortKing-3b-v0.2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/AtAndDev/ShortKing-3b-v0.2

SGLang

How to use AtAndDev/ShortKing-3b-v0.2 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 "AtAndDev/ShortKing-3b-v0.2" \
    --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": "AtAndDev/ShortKing-3b-v0.2",
		"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 "AtAndDev/ShortKing-3b-v0.2" \
        --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": "AtAndDev/ShortKing-3b-v0.2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use AtAndDev/ShortKing-3b-v0.2 with Docker Model Runner:
```
docker model run hf.co/AtAndDev/ShortKing-3b-v0.2
```

Model overview

This model is finetuned on a merged dataset of: oasst1-en, alpaca-cleaned and airoboros-2.1-no-code on a base model: Marx-3b-V2

License: "Creative-Commons-Attribution-4.0"
Language: "en"
Size: "3.43b params"

Prompt template

Prompt template:

### SYSTEM:
<system_prompt_here>

### HUMAN:
<prompter_message_here>

### INPUT:
<input_text_here>

### RESPONSE:
<leave_a_blank_line_here>

Note: If you dont have a system or input text, do not include the tokens in the prompt.

Training Details

This model took 2:40:54 to train in LoRA on a single A100 40gb GPU.

epochs: 1
train batch size: 8
eval batch size: 8
gradient accumulation steps: 1
maximum gradient normal: 0.3
learning rate: 2e-4
weight decay: 0.001
optimizer: paged_adamw_32bit
learning rate schedule: cosine
warmup ratio (linear): 0.03

Downloads last month: 801

Safetensors

Model size

3B params

Tensor type

F32

F16