Instructions to use wenbopan/Faro-Yi-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use wenbopan/Faro-Yi-9B with Transformers:

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

pipe = pipeline("text-generation", model="wenbopan/Faro-Yi-9B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("wenbopan/Faro-Yi-9B")
model = AutoModelForCausalLM.from_pretrained("wenbopan/Faro-Yi-9B")
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 wenbopan/Faro-Yi-9B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/wenbopan/Faro-Yi-9B

SGLang

How to use wenbopan/Faro-Yi-9B 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 "wenbopan/Faro-Yi-9B" \
    --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": "wenbopan/Faro-Yi-9B",
		"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 "wenbopan/Faro-Yi-9B" \
        --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": "wenbopan/Faro-Yi-9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use wenbopan/Faro-Yi-9B with Docker Model Runner:
```
docker model run hf.co/wenbopan/Faro-Yi-9B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

The Faro chat model focuses on practicality and long-context modeling. It handles various downstream tasks with higher quality, delivering stable and reliable results even when inputs contain lengthy documents or complex instructions. Faro seamlessly works in both English and Chinese.

Faro-Yi-9B

Faro-Yi-9B is an improved Yi-9B-200K with extensive instruction tuning on Fusang-V1. Compared to Yi-9B-200K, Faro-Yi-9B has gained greater capability in various downstream tasks and long-context modeling thanks to the large-scale synthetic data in Fusang-V1.

Just like Yi-9B-200K, Faro-Yi-9B supports up to 200K context length.

How to Use

Faro-Yi-9B uses the chatml template and performs well in both short and long contexts. For longer inputs under 24GB of VRAM, I recommend to use vLLM to have a max prompt of 32K. Setting kv_cache_dtype="fp8_e5m2" allows for 48K input length. 4bit-AWQ quantization on top of that can boost input length to 160K, albeit with some performance impact. Adjust max_model_len arg in vLLM or config.json to avoid OOM.

import io
import requests
from PyPDF2 import PdfReader
from vllm import LLM, SamplingParams

llm = LLM(model="wenbopan/Faro-Yi-9B", kv_cache_dtype="fp8_e5m2", max_model_len=100000)

pdf_data = io.BytesIO(requests.get("https://arxiv.org/pdf/2303.08774.pdf").content)
document = "".join(page.extract_text() for page in PdfReader(pdf_data).pages) # 100 pages

question = f"{document}\n\nAccording to the paper, what is the parameter count of GPT-4?"
messages = [ {"role": "user", "content": question} ] # 83K tokens
prompt = llm.get_tokenizer().apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
output = llm.generate(prompt, SamplingParams(temperature=0.8, max_tokens=500))
print(output[0].outputs[0].text)
# Yi-9B-200K:      175B. GPT-4 has 175B \nparameters. How many models were combined to create GPT-4? Answer: 6. ...
# Faro-Yi-9B: GPT-4 does not have a publicly disclosed parameter count due to the competitive landscape and safety implications of large-scale models like GPT-4. ...

Or With Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('wenbopan/Faro-Yi-9B', device_map="cuda")
tokenizer = AutoTokenizer.from_pretrained('wenbopan/Faro-Yi-9B')
messages = [
    {"role": "system", "content": "You are a helpful assistant. Always answer with a short response."},
    {"role": "user", "content": "Tell me what is Pythagorean theorem like you are a pirate."}
]

input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to(model.device)
generated_ids = model.generate(input_ids, max_new_tokens=512, temperature=0.5)
response = tokenizer.decode(generated_ids[0], skip_special_tokens=True) # Aye, matey! The Pythagorean theorem is a nautical rule that helps us find the length of the third side of a triangle. ...

Performance

Faro-Yi-9B enhances its ability compared to Yi-9B-200K in most dimensions, especially in long-range modeling and bilingual (English, Chinese) understanding. Faro is competitive among all open-sourced models at around 9B parameters.

Benchmark Results