Instructions to use IntervitensInc/gemma-2-9b-chatml with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use IntervitensInc/gemma-2-9b-chatml with Transformers:

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

pipe = pipeline("text-generation", model="IntervitensInc/gemma-2-9b-chatml")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("IntervitensInc/gemma-2-9b-chatml")
model = AutoModelForCausalLM.from_pretrained("IntervitensInc/gemma-2-9b-chatml")
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

vLLM

How to use IntervitensInc/gemma-2-9b-chatml with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IntervitensInc/gemma-2-9b-chatml"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IntervitensInc/gemma-2-9b-chatml",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/IntervitensInc/gemma-2-9b-chatml

SGLang

How to use IntervitensInc/gemma-2-9b-chatml 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 "IntervitensInc/gemma-2-9b-chatml" \
    --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": "IntervitensInc/gemma-2-9b-chatml",
		"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 "IntervitensInc/gemma-2-9b-chatml" \
        --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": "IntervitensInc/gemma-2-9b-chatml",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use IntervitensInc/gemma-2-9b-chatml with Docker Model Runner:
```
docker model run hf.co/IntervitensInc/gemma-2-9b-chatml
```

Is this based on instruction tuned gemma or gemma base model?

by jsgreenawalt - opened Sep 29, 2024

Discussion

jsgreenawalt

Sep 29, 2024

I'm getting strange results (nonsense output) when trying to add some fine-tunes based on this model to a merge with gemma-2-9b-it variants, just wanted to confirm this is based on the non-it version of the original Gemma weights?

Thanks!

intervitens

Intervitens AI Innovations Inc. org Sep 30, 2024

This is the same weights as the base model, not -it, not finetuned either, the only modification is replacing two of the reserved tokens in the tokenizer, and changing the eos token in the config.

jsgreenawalt

Oct 2, 2024

Thanks for the info

jsgreenawalt changed discussion status to closed Oct 2, 2024

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment