Instructions to use manasdutta04/kudu-writer-tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use manasdutta04/kudu-writer-tiny 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 manasdutta04/kudu-writer-tiny:IQ4_XS
# Run inference directly in the terminal:
llama cli -hf manasdutta04/kudu-writer-tiny:IQ4_XS

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf manasdutta04/kudu-writer-tiny:IQ4_XS
# Run inference directly in the terminal:
llama cli -hf manasdutta04/kudu-writer-tiny:IQ4_XS

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 manasdutta04/kudu-writer-tiny:IQ4_XS
# Run inference directly in the terminal:
./llama-cli -hf manasdutta04/kudu-writer-tiny:IQ4_XS

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 manasdutta04/kudu-writer-tiny:IQ4_XS
# Run inference directly in the terminal:
./build/bin/llama-cli -hf manasdutta04/kudu-writer-tiny:IQ4_XS

Use Docker

docker model run hf.co/manasdutta04/kudu-writer-tiny:IQ4_XS

Llama new
LM Studio
Jan

vLLM

How to use manasdutta04/kudu-writer-tiny with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "manasdutta04/kudu-writer-tiny"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "manasdutta04/kudu-writer-tiny",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/manasdutta04/kudu-writer-tiny:IQ4_XS

Ollama
How to use manasdutta04/kudu-writer-tiny with Ollama:
```
ollama run hf.co/manasdutta04/kudu-writer-tiny:IQ4_XS
```
Unsloth Desktop
Docker Model Runner
How to use manasdutta04/kudu-writer-tiny with Docker Model Runner:
```
docker model run hf.co/manasdutta04/kudu-writer-tiny:IQ4_XS
```

Lemonade

How to use manasdutta04/kudu-writer-tiny with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull manasdutta04/kudu-writer-tiny:IQ4_XS

Run and chat with the model

lemonade run user.kudu-writer-tiny-IQ4_XS

List all available models

lemonade list

Atomic Chat

kudu-writer-tiny

Compact on-device text formatter for voice dictation.

Benchmarks

Model	Replies accepted	Exact match	Latency (50 words)
kudu-writer-tiny, short prompt	98.5%	70.0%	0.87 s
kudu-writer-tiny, Kudu's current prompt	98.9%	69.6%	1.72 s
Qwen2.5-0.5B-Instruct (stock)	79.8%	9.6%	2.75 s
SmolLM2-360M-Instruct (stock)	62.5%	13.3%	2.96 s

Internal benchmark: 270 dictation cases, CPU only (4 threads).

License

Apache-2.0. Fine-tuned from SmolLM2-360M-Instruct (Apache-2.0).

Downloads last month: 127

GGUF

Model size

0.4B params

Architecture

llama

Hardware compatibility

4-bit

Model tree for manasdutta04/kudu-writer-tiny

Base model

HuggingFaceTB/SmolLM2-360M

Quantized

HuggingFaceTB/SmolLM2-360M-Instruct

Quantized

(132)

this model