Instructions to use yys/gemma-7B-it-firefly with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yys/gemma-7B-it-firefly with Transformers:

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

pipe = pipeline("text-generation", model="yys/gemma-7B-it-firefly")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("yys/gemma-7B-it-firefly")
model = AutoModelForCausalLM.from_pretrained("yys/gemma-7B-it-firefly", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yys/gemma-7B-it-firefly with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yys/gemma-7B-it-firefly"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yys/gemma-7B-it-firefly",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yys/gemma-7B-it-firefly

SGLang

How to use yys/gemma-7B-it-firefly 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 "yys/gemma-7B-it-firefly" \
    --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": "yys/gemma-7B-it-firefly",
		"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 "yys/gemma-7B-it-firefly" \
        --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": "yys/gemma-7B-it-firefly",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use yys/gemma-7B-it-firefly with Docker Model Runner:
```
docker model run hf.co/yys/gemma-7B-it-firefly
```

Model Card for Firefly-Gemma

gemma-7B-it-firefly is trained based on gemma-7b-it to act as a helpful and harmless AI assistant. we trained the model on firefly-train-1.1M dataset using LoRA.

Performance

we evaluate the model on Open LLM Leaderboard

Usage

The chat template of our chat models is same as Official gemma-7b-it:

<bos><start_of_turn>user
Write a hello world program<end_of_turn>
<start_of_turn>model

You can also use the following code:

from transformers import AutoTokenizer, AutoModelForCausalLM

model_name_or_path = "yys/gemma-7B-it-firefly"
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
model = AutoModelForCausalLM.from_pretrained(model_name_or_path)

input_text = "给我写一首关于机器学习的诗歌。"
input_ids = tokenizer(input_text, return_tensors="pt")

outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))

Downloads last month: 55

Safetensors

Model size

9B params

Tensor type

BF16