Instructions to use kroffske/shellcue-lfm2.5-230m-alpha with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kroffske/shellcue-lfm2.5-230m-alpha with Transformers:

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

pipe = pipeline("text-generation", model="kroffske/shellcue-lfm2.5-230m-alpha")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kroffske/shellcue-lfm2.5-230m-alpha")
model = AutoModelForCausalLM.from_pretrained("kroffske/shellcue-lfm2.5-230m-alpha", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use kroffske/shellcue-lfm2.5-230m-alpha with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kroffske/shellcue-lfm2.5-230m-alpha"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kroffske/shellcue-lfm2.5-230m-alpha",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/kroffske/shellcue-lfm2.5-230m-alpha

SGLang

How to use kroffske/shellcue-lfm2.5-230m-alpha 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 "kroffske/shellcue-lfm2.5-230m-alpha" \
    --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": "kroffske/shellcue-lfm2.5-230m-alpha",
		"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 "kroffske/shellcue-lfm2.5-230m-alpha" \
        --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": "kroffske/shellcue-lfm2.5-230m-alpha",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use kroffske/shellcue-lfm2.5-230m-alpha with Docker Model Runner:
```
docker model run hf.co/kroffske/shellcue-lfm2.5-230m-alpha
```

ShellCue LFM2.5 230M Alpha

Development alpha with DEV_GRADE_SUPPORT for local shell-command suffix completion. This snapshot is not product-accepted, trusted, final, or production-ready. It predicts a short suffix from masked recent command context and a typed prefix; it does not execute commands.

Runtime

Use ShellCue >=0.1.0a1. Download and install this model by an immutable Hugging Face commit OID, then run shellcue model verify <snapshot-directory> before use. The readable v0.1.0a1 tag is a non-moving alias, not the trust anchor.

Origin and license

This model is a full fine-tune of LiquidAI/LFM2.5-230M-Base at revision 9d2be5519834990d30996f878b6771cccbd24f2c. The weights are distributed under the LFM Open License v1.0 in LICENSE, not the MIT license used by the separate ShellCue runtime code. See NOTICE for the modification notice and attribution. The license includes a commercial-use limitation; review its complete terms before use or redistribution.

Limitations

Development evidence does not establish quality or safety for every shell session.
Suggestions require user review and may be incorrect, incomplete, or unsafe.
No hosted inference, telemetry, or implicit network download is part of runtime use.

Integrity

model.safetensors SHA-256: c4f7973c48eb04fa2e8013f0d03171fcfb4ee27c157dea31e96020b12b84fb53
Base revision: 9d2be5519834990d30996f878b6771cccbd24f2c
Release contract: shellcue.model.v1 and shellcue.inference.v1

Downloads last month: 209

Safetensors

Model size

0.2B params

Tensor type

F32

Model tree for kroffske/shellcue-lfm2.5-230m-alpha

Base model

LiquidAI/LFM2.5-230M-Base

Finetuned

(8)

this model