Instructions to use nroggendorff/smallama-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nroggendorff/smallama-7b with Transformers:

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

pipe = pipeline("text-generation", model="nroggendorff/smallama-7b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nroggendorff/smallama-7b")
model = AutoModelForCausalLM.from_pretrained("nroggendorff/smallama-7b", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nroggendorff/smallama-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nroggendorff/smallama-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nroggendorff/smallama-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/nroggendorff/smallama-7b

SGLang

How to use nroggendorff/smallama-7b 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 "nroggendorff/smallama-7b" \
    --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": "nroggendorff/smallama-7b",
		"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 "nroggendorff/smallama-7b" \
        --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": "nroggendorff/smallama-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nroggendorff/smallama-7b with Docker Model Runner:
```
docker model run hf.co/nroggendorff/smallama-7b
```

nroggendorff commited on Oct 10, 2025

Commit

e7014e4

verified ·

1 Parent(s): b4f62b1

Training loss: 1.3980

Browse files

Files changed (3) hide show

model-00001-of-00003.safetensors +1 -1
model-00002-of-00003.safetensors +1 -1
model-00003-of-00003.safetensors +1 -1

model-00001-of-00003.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a60e0cba5e0803957dfb822bb6d4e7312cd91f767f1c84e9f1353e669218df51
 size 4994007360

 version https://git-lfs.github.com/spec/v1
+oid sha256:b1160048809f1cec3c0417dfbbca56e8d72e9c99ea0aaf48ba75bae9e92d17e2
 size 4994007360

model-00002-of-00003.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36446f022e9ec1a593198271a667876131357f3c18df94cb99a15ad919e99ba4
 size 4925194872

 version https://git-lfs.github.com/spec/v1
+oid sha256:c6bc535433200e71b6ec943dbaac4694c49790b063b88544154e341b5734f1e0
 size 4925194872

model-00003-of-00003.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e4cb5e28cfc4b19564efd3afe4b52c6d1a1da79d22c32bf784c0945439e13f25
 size 4728593704

 version https://git-lfs.github.com/spec/v1
+oid sha256:a97a43c37d4f48478c60d55d7fa696c13eb05ef8d8adf961dbb40b4823755c14
 size 4728593704