Instructions to use abacusai/Smaug-Mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use abacusai/Smaug-Mini with Transformers:

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

pipe = pipeline("image-text-to-text", model="abacusai/Smaug-Mini")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("abacusai/Smaug-Mini")
model = AutoModelForMultimodalLM.from_pretrained("abacusai/Smaug-Mini", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use abacusai/Smaug-Mini with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "abacusai/Smaug-Mini"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "abacusai/Smaug-Mini",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/abacusai/Smaug-Mini

SGLang

How to use abacusai/Smaug-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 "abacusai/Smaug-Mini" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "abacusai/Smaug-Mini",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "abacusai/Smaug-Mini" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "abacusai/Smaug-Mini",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use abacusai/Smaug-Mini with Docker Model Runner:
```
docker model run hf.co/abacusai/Smaug-Mini
```

prann-ab commited on Sep 3

Commit

e5d9f2b

verified ·

1 Parent(s): 159f1cb

Drop redundant eval tables; regenerate vs-base chart with matching font scale

Browse files

Files changed (3) hide show

.gitattributes +1 -0
README.md +0 -16
figures/smaug_mini_vs_base.png +0 -0

.gitattributes CHANGED Viewed

@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
 figures/smaug_mini_livebench_categories.png filter=lfs diff=lfs merge=lfs -text

 *tfevents* filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
 figures/smaug_mini_livebench_categories.png filter=lfs diff=lfs merge=lfs -text
+figures/smaug_mini_vs_base.png filter=lfs diff=lfs merge=lfs -text

README.md CHANGED Viewed

@@ -51,24 +51,8 @@ One deployment note: the multi-token-prediction (MTP) head is inherited from the
 All Smaug-Mini numbers below were produced at temperature 1.0, top_p 0.95, reasoning effort `xhigh`, with generation budgets large enough that no score is truncation-bound. Base-model comparison points are the published Qwen3.8-27B model-card numbers and the public [livebench.ai](https://livebench.ai) leaderboard entries, never our own reruns of the base.
-### Agentic
-| Benchmark | Smaug-Mini | Qwen3.8-27B | Delta |
-|---|---|---|---|
-| AutomationBench (full 600) | **41.8** | 37.3 | +4.5 |
-| JobBench (macro) | **50.5** | 33.4 | +17.1 |
-| LiveBench agentic coding | 61.4 | 61.4 | par |
 JobBench was run under the official protocol with the OpenCode scaffold and an LLM judge; the base-model comparison point is the score reported on the Qwen3.8-27B model card.
-### Reasoning, knowledge and instruction following
-| Benchmark | Smaug-Mini | Qwen3.8-27B | Delta |
-|---|---|---|---|
-| GPQA-diamond | **89.4** | 89.2 | +0.2 |
-| IFBench (prompt-level loose) | **82.0** | 79.5 | +2.5 |
-| LiveBench overall | **77.2** | 75.3 | +2.0 |
 ### LiveBench by category (vs published leaderboard)
 <div align="center">

 All Smaug-Mini numbers below were produced at temperature 1.0, top_p 0.95, reasoning effort `xhigh`, with generation budgets large enough that no score is truncation-bound. Base-model comparison points are the published Qwen3.8-27B model-card numbers and the public [livebench.ai](https://livebench.ai) leaderboard entries, never our own reruns of the base.
 JobBench was run under the official protocol with the OpenCode scaffold and an LLM judge; the base-model comparison point is the score reported on the Qwen3.8-27B model card.
 ### LiveBench by category (vs published leaderboard)
 <div align="center">

figures/smaug_mini_vs_base.png CHANGED Viewed

Git LFS Details

SHA256: 20dbcd529ba4282742dbb17a1453cab53fb77b2cfe43eaaf70082f1ef5ca48f8
Pointer size: 131 Bytes
Size of remote file: 209 kB