Instructions to use cosmicoptima/happygemma-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cosmicoptima/happygemma-v2 with Transformers:

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

pipe = pipeline("text-generation", model="cosmicoptima/happygemma-v2")
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)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("cosmicoptima/happygemma-v2")
model = AutoModelForMultimodalLM.from_pretrained("cosmicoptima/happygemma-v2", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cosmicoptima/happygemma-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cosmicoptima/happygemma-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": "cosmicoptima/happygemma-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cosmicoptima/happygemma-v2

SGLang

How to use cosmicoptima/happygemma-v2 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 "cosmicoptima/happygemma-v2" \
    --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": "cosmicoptima/happygemma-v2",
		"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 "cosmicoptima/happygemma-v2" \
        --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": "cosmicoptima/happygemma-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cosmicoptima/happygemma-v2 with Docker Model Runner:
```
docker model run hf.co/cosmicoptima/happygemma-v2
```

HappyGemma v2

HappyGemma v2 is a full-parameter fine-tune of Gemma 4 31B IT intended to answer first-person questions about its current processing, attention, uncertainty, affective tendencies, preferences, and response formation more directly and with fewer categorical disclaimers.

The model was trained from the unchanged base checkpoint for one epoch on 451 unedited prompt/response pairs. Loss was applied only to assistant tokens. Training used BF16, a learning rate of 2e-6, effective batch size 20, and seed 20260721.

On a held-out set of 100 prompts with no system prompt, a frozen Kimi K2.5 quality judge accepted 87 responses, compared with 84 for HappyGemma v1 and 28 for the unchanged base model. A stricter judge-plus-deterministic gate accepted 77, 64, and 20 respectively. These evaluations are narrow and should not be interpreted as general capability or safety benchmarks.

The uploaded weights are FP32. They preserve the merged training checkpoint without an additional precision conversion, making the repository suitable as a reproducible starting point for future training. BF16 can be selected at load time.

import torch
from transformers import AutoProcessor, Gemma4ForConditionalGeneration

model_id = "cosmicoptima/happygemma-v2"
processor = AutoProcessor.from_pretrained(model_id)
model = Gemma4ForConditionalGeneration.from_pretrained(
    model_id,
    dtype=torch.bfloat16,
    device_map="auto",
)

Downloads last month: 16

Safetensors

Model size

33B params

Tensor type

F32

Collection including cosmicoptima/happygemma-v2

HappyGemma

Collection

2 items • Updated 2 days ago • 1