Instructions to use SeacomSrl/SeaPhi3-medium with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SeacomSrl/SeaPhi3-medium with Transformers:

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

pipe = pipeline("text-generation", model="SeacomSrl/SeaPhi3-medium")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SeacomSrl/SeaPhi3-medium")
model = AutoModelForCausalLM.from_pretrained("SeacomSrl/SeaPhi3-medium")
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 SeacomSrl/SeaPhi3-medium with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SeacomSrl/SeaPhi3-medium"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SeacomSrl/SeaPhi3-medium",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SeacomSrl/SeaPhi3-medium

SGLang

How to use SeacomSrl/SeaPhi3-medium 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 "SeacomSrl/SeaPhi3-medium" \
    --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": "SeacomSrl/SeaPhi3-medium",
		"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 "SeacomSrl/SeaPhi3-medium" \
        --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": "SeacomSrl/SeaPhi3-medium",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SeacomSrl/SeaPhi3-medium with Docker Model Runner:
```
docker model run hf.co/SeacomSrl/SeaPhi3-medium
```

Uploaded model

Developed by: Toti Riccardo
License: apache-2.0
Finetuned from model : unsloth/Phi-3-medium-4k-instruct on Seacom/rag-data an italian translated dataset.

Evaluation

For a detailed comparison of model performance, check out the Leaderboard for Italian Language Models.

Here's a breakdown of the performance metrics:

Metric	hellaswag_it acc_norm	arc_it acc_norm	m_mmlu_it 5-shot acc	m_mmlu_pro_it	Average
Accuracy Normalized	69.35	49.1	69.53	0.3913	62.66

Downloads last month: 2,650

Safetensors

Model size

14B params

Tensor type

F32

F16

Dataset used to train SeacomSrl/SeaPhi3-medium

Collection including SeacomSrl/SeaPhi3-medium

FULL LLM

Collection

Una collezione di LLM fine-tuned sul nostro dataset • 4 items • Updated Jun 7, 2024 • 1