Instructions to use google/recurrentgemma-2b-it with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use google/recurrentgemma-2b-it with Transformers:

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

pipe = pipeline("text-generation", model="google/recurrentgemma-2b-it")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/recurrentgemma-2b-it")
model = AutoModelForCausalLM.from_pretrained("google/recurrentgemma-2b-it")
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 google/recurrentgemma-2b-it with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/google/recurrentgemma-2b-it

SGLang

How to use google/recurrentgemma-2b-it 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 "google/recurrentgemma-2b-it" \
    --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": "google/recurrentgemma-2b-it",
		"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 "google/recurrentgemma-2b-it" \
        --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": "google/recurrentgemma-2b-it",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use google/recurrentgemma-2b-it with Docker Model Runner:
```
docker model run hf.co/google/recurrentgemma-2b-it
```

Can this thing be quantized?

by Winmodel - opened Apr 10, 2024

Discussion

Winmodel

Apr 10, 2024

Can this thing be quantized? or its recurrent block is sensitive to low precision like Mamba?

ybelkada

Apr 10, 2024

Hi @Winmodel
I think as long as the model contains torch.nn.Linear layers it can be quantized with any of these methods: https://huggingface.co/docs/transformers/quantization

slsmith

Google org Apr 10, 2024

We haven't investigated quantizing RecurrentGemma/Griffin models yet.
I would expect that the linear layers (which has most of the weights) can be quantized, but it might be harder to quantize the recurrent layer.

Note that for the recurrence, we recommend storing the parameters in bfloat16, but casting them on device to float32 to perform the recurrence in high precision.
Casting the parameters to float32 on device is free since the recurrence is memory bound, and it significantly improves numerical precision.

sdalemorrey

Apr 16, 2024

Thanks for this great model! Is there a gguf version of this yet?

gmuraru

Apr 23, 2024

Hello @sdalemorrey .

We do not have a gguf version of the model and at the moment we do not plan to release one.

In case it might be helpful, it looks like someone did a version of a quantized model here[1].
NB: This was not released by us and we did not test or evaluate it.

[1] https://huggingface.co/PrunaAI/recurrentgemma-2b-it-bnb-4bit-smashed

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