Instructions to use arcee-ai/raspberry-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use arcee-ai/raspberry-3B with Transformers:

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

pipe = pipeline("text-generation", model="arcee-ai/raspberry-3B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("arcee-ai/raspberry-3B")
model = AutoModelForCausalLM.from_pretrained("arcee-ai/raspberry-3B")
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]:]))

Inference
Local Apps Settings

vLLM

How to use arcee-ai/raspberry-3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "arcee-ai/raspberry-3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "arcee-ai/raspberry-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/arcee-ai/raspberry-3B

SGLang

How to use arcee-ai/raspberry-3B 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 "arcee-ai/raspberry-3B" \
    --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": "arcee-ai/raspberry-3B",
		"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 "arcee-ai/raspberry-3B" \
        --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": "arcee-ai/raspberry-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use arcee-ai/raspberry-3B with Docker Model Runner:
```
docker model run hf.co/arcee-ai/raspberry-3B
```

Exploring Raspberry-3B: Use Cases, Performance, and Optimization

by Salkic - opened Oct 7, 2024

Discussion

Salkic

Oct 7, 2024

Hi everyone!

This is my first time posting here, and I’m really excited to learn from the community. I’ve been exploring the Raspberry-3B LLM model and wanted to start a discussion about its capabilities and how others are using it.

I was amazed to see that it’s currently beating Llama3.1-8B on the leaderboard. I thought it would be interesting to discuss how this smaller model is achieving such performance.

Salkic changed discussion status to closed Oct 7, 2024

Salkic changed discussion status to open Oct 7, 2024

Antix5

Oct 14, 2024

Honestly, I'm not convinced, the model is finetuned for reasoning but for even really simple reasoning, it's not even coherent in a very simple math question, I know it's a 3B but it's unusable (tried with the quantized version, might be ok with the full version):
User : A > B > C > D, A,B,C,D in R, is this statement true D < A
LLM : B < C)
Certainly not. That's how I understand it. "It's that extra_id_1 the user doesn't like to be on their own.") It was a nice way of saying they don't like each other.
A > B > C > D).I think you mean in this context, like A and C are people or entities who share a common goal or ideology but it might not necessarily apply to individuals.A person could share an
idea with another person that doesn't necessarily mean the two people personally dislike each other -Surely not every individual dislikes everyone else]).

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