Instructions to use KittyLM/kittylm-gemma3-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KittyLM/kittylm-gemma3-4b with Transformers:

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

pipe = pipeline("text-generation", model="KittyLM/kittylm-gemma3-4b")
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("KittyLM/kittylm-gemma3-4b")
model = AutoModelForMultimodalLM.from_pretrained("KittyLM/kittylm-gemma3-4b", 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 KittyLM/kittylm-gemma3-4b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KittyLM/kittylm-gemma3-4b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KittyLM/kittylm-gemma3-4b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/KittyLM/kittylm-gemma3-4b

SGLang

How to use KittyLM/kittylm-gemma3-4b 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 "KittyLM/kittylm-gemma3-4b" \
    --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": "KittyLM/kittylm-gemma3-4b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "KittyLM/kittylm-gemma3-4b" \
        --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": "KittyLM/kittylm-gemma3-4b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use KittyLM/kittylm-gemma3-4b with Docker Model Runner:
```
docker model run hf.co/KittyLM/kittylm-gemma3-4b
```

Lalaggi commited on 13 days ago

Commit

a3ffb63

verified ·

1 Parent(s): 86d2955

model card

Browse files

Files changed (1) hide show

README.md +58 -0

README.md ADDED Viewed

	@@ -0,0 +1,58 @@

+---
+library_name: transformers
+pipeline_tag: text-generation
+license: gemma
+base_model: google/gemma-3-4b-it
+datasets: KittyLM/kittylm-data
+tags:
+- kittylm
+- lora
+- persona
+- roleplay
+widget:
+- messages:
+  - role: user
+    content: Hey!
+---
+# KittyLM-4B (gemma3-4b)
+A kitten. In a language model. This is a LoRA finetune of [google/gemma-3-4b-it](https://huggingface.co/google/gemma-3-4b-it) that answers **everything** in kitten language
+(mrrp, nya~, prrr, `*actions*`, occasional `:3`) while staying factually correct underneath.
+> **License note:** this is a derivative of a Gemma model and is distributed under the [Gemma Terms of Use](https://ai.google.dev/gemma/terms). You must accept Google's license on HuggingFace before downloading.
+## Training
+- Data: 900 ShareGPT-style pairs + 100 held-out eval (see `KittyLM/kittylm-data`), system prompt baked in
+- Method: LoRA SFT (`scripts/train.py` in the KittyLM project), RTX 3060 12GB
+- Files: merged bf16 weights + the LoRA adapter (`adapter_*.safetensors`) live side by side.
+  `AutoModelForCausalLM` loads the merged model; `PeftModel` picks up the adapter.
+- GGUF quants for Ollama / llama.cpp / LM Studio: [KittyLM/kittylm-gemma3-4b-gguf](https://huggingface.co/KittyLM/kittylm-gemma3-4b-gguf)
+- Eval: v1 (prompt-free): ablation full/none/generic style 0.75/0.75/0.70, 15/15 factual. Thinnest large model but passes both bars. Gemma3 multimodal: image_soft_token excluded from GGUF (text-only deploy). Train loss 7.9->0.36, eval loss 0.850->0.846->0.952 (best @ epoch 2).
+## Limitations
+- Persona is stylistic, not a refusal behavior: an explicit "answer in plain English" can make it drop character — it was never trained to resist.
+- Small-model knowledge gaps persist: off-distribution facts may confabulate (tracked per model by the 20-probe suite in `eval/`).
+- Kitten flavor adds no capability: reasoning/coding limits are the base model's limits.
+## Usage (transformers)
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+tok = AutoTokenizer.from_pretrained("KittyLM/kittylm-gemma3-4b")
+model = AutoModelForCausalLM.from_pretrained("KittyLM/kittylm-gemma3-4b", device_map="auto", dtype="bfloat16")
+msgs = [{"role": "user", "content": "What's the capital of Japan?"}]
+x = tok.apply_chat_template(msgs, tokenize=False, add_generation_prompt=True)
+print(tok.decode(model.generate(**tok(x, return_tensors="pt").to(model.device), max_new_tokens=120)[0]))
+# mrrp... Tokyo. big city. lots of cats. nya~
+```
+## Usage (Ollama)
+```bash
+hf download KittyLM/kittylm-gemma3-4b-gguf --include "*.gguf" --local-dir ./gguf
+ollama create kittylm-gemma3-4b -f Modelfile   # see Modelfile template in project
+ollama run kittylm-gemma3-4b "Good night!"
+```