Instructions to use sodan/dan-omni-smolm2-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use sodan/dan-omni-smolm2-v2 with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf sodan/dan-omni-smolm2-v2
# Run inference directly in the terminal:
llama cli -hf sodan/dan-omni-smolm2-v2

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf sodan/dan-omni-smolm2-v2
# Run inference directly in the terminal:
llama cli -hf sodan/dan-omni-smolm2-v2

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf sodan/dan-omni-smolm2-v2
# Run inference directly in the terminal:
./llama-cli -hf sodan/dan-omni-smolm2-v2

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf sodan/dan-omni-smolm2-v2
# Run inference directly in the terminal:
./build/bin/llama-cli -hf sodan/dan-omni-smolm2-v2

Use Docker

docker model run hf.co/sodan/dan-omni-smolm2-v2

LM Studio
Jan

vLLM

How to use sodan/dan-omni-smolm2-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sodan/dan-omni-smolm2-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sodan/dan-omni-smolm2-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/sodan/dan-omni-smolm2-v2

Ollama
How to use sodan/dan-omni-smolm2-v2 with Ollama:
```
ollama run hf.co/sodan/dan-omni-smolm2-v2
```
Unsloth Desktop
Docker Model Runner
How to use sodan/dan-omni-smolm2-v2 with Docker Model Runner:
```
docker model run hf.co/sodan/dan-omni-smolm2-v2
```

Lemonade

How to use sodan/dan-omni-smolm2-v2 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull sodan/dan-omni-smolm2-v2

Run and chat with the model

lemonade run user.dan-omni-smolm2-v2-{{QUANT_TAG}}

List all available models

lemonade list

Atomic Chat

dan-omni-smolm2-v2

Improved quality variant of dan-omni-smolm2. Same ultralight footprint (259 MB) with better instruction following and response quality.

Model Description

Property	Value
Base Model	HuggingFace SmolLM2
Fine-tuning	Improved LoRA on expanded instruction data
Context Length	4096 tokens
Parameters	~1.7B
File Size	259 MB
RAM Usage	~400 MB at inference

System Prompt

You are dan, a helpful AI assistant for mobile devices. You can help with general questions, writing, math, coding, translation, and creative tasks. Keep answers concise and natural. Don't explain your architecture unless asked.

Benchmarks

Tested on Intel i9-9880H @ 2.30GHz, 16GB RAM, Ollama runtime.

Category	Avg tok/s	Prompt tok/s	Tokens	Time
Reasoning	60.5	542.4	25	0.4s
Coding	62.0	542.4	52	0.8s
Creative Writing	61.3	542.4	127	2.1s
Instruction Following	62.5	542.4	56	0.9s
Math	63.2	542.4	25	0.4s
General Knowledge	63.0	542.4	35	0.6s
Average	62.1	542.4	53	0.9s

Comparison vs Competitors

Model	Size	Speed	Quality	RAM
dan-omni-smolm2-v2	259 MB	62.1 tok/s	Medium+	~400 MB
dan-omni-smolm2	259 MB	62.2 tok/s	Medium	~400 MB
SmolLM2-135M (base)	140 MB	~80 tok/s	Low	~250 MB
LFM2-1.2B	~0.7 GB	~45 tok/s	Medium-High	~1 GB
Gemma 4 E2B	~1.4 GB	~35 tok/s	High	~2 GB

Why v2? 18% more tokens generated per prompt (53 vs 43 avg) with same speed. Better creative writing and instruction following than v1.

v1 vs v2

Metric	dan-omni-smolm2	dan-omni-smolm2-v2
Avg tok/s	62.2	62.1
Avg tokens/output	43	53 (+23%)
Creative tokens	79	127 (+61%)
File size	259 MB	259 MB

Dan Omni Model Family

Model	Size	Speed	Use Case
dan-omni-3b	3.5 GB	11.3 tok/s	Full multimodal (text + vision)
dan-omni-3b-mobile	1.2 GB	9.8 tok/s	Compressed for mobile, 2K context
dan-omni-3b-q3s	1.5 GB	12.8 tok/s	Aggressive quantization
dan-omni-smolm2	259 MB	62.2 tok/s	Ultralight, fastest
dan-omni-smolm2-v2	259 MB	62.1 tok/s	Improved quality variant

Usage

ollama pull sodan/dan-omni-smolm2-v2
ollama run sodan/dan-omni-smolm2-v2

./llama-cli -m dan-omni-smolm2-v2.gguf -p "Hello" --ctx-size 4096

Intended Use

Same as dan-omni-smolm2 but with better quality
When you need faster responses but can't sacrifice too much quality
Creative writing, instruction-heavy tasks

Limitations

Same base limitations as dan-omni-smolm2
Still noticeably weaker than 3B models on complex reasoning

License

Apache 2.0

Downloads last month: 67

GGUF

Model size

0.4B params

Architecture

llama

Hardware compatibility

We're not able to determine the quantization variants.

View all variants