Instructions to use c2pig/terminal-shell-mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use c2pig/terminal-shell-mini with Transformers:

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

pipe = pipeline("text-generation", model="c2pig/terminal-shell-mini")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("c2pig/terminal-shell-mini")
model = AutoModelForCausalLM.from_pretrained("c2pig/terminal-shell-mini", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use c2pig/terminal-shell-mini with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "c2pig/terminal-shell-mini"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "c2pig/terminal-shell-mini",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/c2pig/terminal-shell-mini

SGLang

How to use c2pig/terminal-shell-mini 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 "c2pig/terminal-shell-mini" \
    --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": "c2pig/terminal-shell-mini",
		"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 "c2pig/terminal-shell-mini" \
        --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": "c2pig/terminal-shell-mini",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use c2pig/terminal-shell-mini with Docker Model Runner:
```
docker model run hf.co/c2pig/terminal-shell-mini
```

terminal-shell-mini

Tiny GPT-2-style model that imitates Ubuntu 22.04 bash output and answers short questions about a session.


Parameters	3353088
Git SHA	`47476878a75b3c7627704a60d6d0dc34ac2b8d23`
Tokenizer	byte-level BPE fit on the training text, plus chat markers
Training	GitHub Actions CPU runner, `ml/config/tiny.yaml`
Data	Synthetic `ubuntu:22.04` sessions from `ml/data/generate_sessions.py`

Prompt format

<|system|>
You are a Linux terminal on Ubuntu. ...
<|user|>
user@dev:~$ pwd
<|assistant|>
/home/user
<|end|>

The local server in serve/ turns OpenAI messages into this prompt.

Intended use

A toy terminal for terminal-gpt pointed at POST /v1/chat/completions. It is not a real shell, and it is not a general chatbot.

How this checkpoint was published

GitHub Actions workflow ML train uploaded this folder with HF_TOKEN. The git SHA above is the commit that produced the weights.

Downloads last month: 307

Safetensors

Model size

3.35M params

Tensor type

F32

c2pig
/

terminal-shell-mini

terminal-shell-mini

Prompt format

Intended use

How this checkpoint was published

Space using c2pig/terminal-shell-mini 1