llm.create_chat_completion(
messages = "\"The tower is 324 metres (1,063 ft) tall, about the same height as an 81-storey building, and the tallest structure in Paris. Its base is square, measuring 125 metres (410 ft) on each side. During its construction, the Eiffel Tower surpassed the Washington Monument to become the tallest man-made structure in the world, a title it held for 41 years until the Chrysler Building in New York City was finished in 1930. It was the first structure to reach a height of 300 metres. Due to the addition of a broadcasting aerial at the top of the tower in 1957, it is now taller than the Chrysler Building by 5.2 metres (17 ft). Excluding transmitters, the Eiffel Tower is the second tallest free-standing structure in France after the Millau Viaduct.\""
)catnip-summarizer
Fine-tuned Gemma-3-270M for task summarization and branch naming. It's meant to be used summarize user requests to coding agents. For example:
ollama run --verbose hf.co/vanpelt/catnip-summarizer "Help me make the vibes good"
Make Good Vibes
feat/make-good-vibes
The summary and git branch name can be used by tools like Catnip to keep multiple coding agent sessions organized.
Model Details
- Base Model: google/gemma-3-270m-it
- Format: GGUF
- Quantization: Q4_K_M
- Use Case: Generating concise task titles and git branch names
Training
- Training Run: https://wandb.ai/vanpelt/summarizer/runs/jh27ahps
- Repo: https://github.com/vanpelt/summarizer
Usage
Web Demo
You can try the model in your browser!
With Ollama
ollama run hf.co/vanpelt/catnip-summarizer "Help me make the vibes good"
With llama.cpp
# Download the GGUF file
huggingface-cli download vanpelt/catnip-summarizer gemma3-270m-summarizer-Q4_K_M.gguf
# Run with llama.cpp
./main -m gemma3-270m-summarizer-Q4_K_M.gguf -p 'Your prompt here'
Files
tokenizer.json(31.8 MB)tokenizer_config.json(1.1 MB)added_tokens.json(0.0 MB)chat_template.jinja(0.0 MB)Modelfile(0.0 MB)template(0.0 MB)system(0.0 MB)gemma3-270m-summarizer-Q4_K_M.gguf(241.4 MB)special_tokens_map.json(0.0 MB)config.json(0.0 MB)params(0.0 MB)tokenizer.model(4.5 MB)
- Downloads last month
- 8
Hardware compatibility
Log In to add your hardware
4-bit
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="vanpelt/catnip-summarizer", filename="gemma3-270m-summarizer-Q4_K_M.gguf", )