Instructions to use PetroGPT/Breeze-Petro-7B-Instruct-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use PetroGPT/Breeze-Petro-7B-Instruct-v1 with Transformers:

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

pipe = pipeline("text-generation", model="PetroGPT/Breeze-Petro-7B-Instruct-v1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("PetroGPT/Breeze-Petro-7B-Instruct-v1")
model = AutoModelForCausalLM.from_pretrained("PetroGPT/Breeze-Petro-7B-Instruct-v1", device_map="auto")
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 PetroGPT/Breeze-Petro-7B-Instruct-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "PetroGPT/Breeze-Petro-7B-Instruct-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PetroGPT/Breeze-Petro-7B-Instruct-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/PetroGPT/Breeze-Petro-7B-Instruct-v1

SGLang

How to use PetroGPT/Breeze-Petro-7B-Instruct-v1 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 "PetroGPT/Breeze-Petro-7B-Instruct-v1" \
    --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": "PetroGPT/Breeze-Petro-7B-Instruct-v1",
		"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 "PetroGPT/Breeze-Petro-7B-Instruct-v1" \
        --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": "PetroGPT/Breeze-Petro-7B-Instruct-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use PetroGPT/Breeze-Petro-7B-Instruct-v1 with Docker Model Runner:
```
docker model run hf.co/PetroGPT/Breeze-Petro-7B-Instruct-v1
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Breeze-Petro-7B-Instruct-v1

Model creator: MediaTek Research
Original model: MediaTek-Research/Breeze-7B-Instruct-v1_0

Model Description

This is a model fine-tuned based on Breeze-7B-Instruct-v1_0. The training set is mainly based on chemical knowledge and procedural knowledge. Supplement knowledge about the petroleum industry.

Developed by: RebeccaChou
License: apache-2.0
Finetuned from model : [MediaTek-Research/Breeze-7B-Instruct-v1_0
Language(s) (NLP): [English.繁體中文]

📖 Table of Contents

1.Open LLM Leaderboard - ARC - HellaSwag - MMLU - TruthfulQA - Winogrande - GSM8K 3. EvalPlus Leaderboard - HumanEval - HumanEval_Plus - MBPP - MBPP_Plus 4. Prompt Format 5. Quantized Models 6. Gratitude

🏆 Open LLM Leaderboard

WestSeverus-7B-DPO-v2 is one of the top 7B model in Open LLM Leaderboard and it outperforms on TruthfulQA and GSM8K.

Metric	Value
Avg.	59.32
AI2 Reasoning Challenge (25-Shot)	58.87
HellaSwag (10-Shot)	79.17
MMLU (5-Shot)	56.62
TruthfulQA (0-shot)	46.36
Winogrande (5-shot)	73.64
GSM8k (5-shot)	41.24

Detailed results can be found here

⚡ EvalPlus Leaderboard

Model	HumanEval	HumanEval_Plus	MBPP	MBPP_Plus
phi-2-2.7B	48.2	43.3	61.9	51.4

SOLAR-10.7B-Instruct-v1.0	42.1	34.3	42.9	34.6
CodeLlama-7B	37.8	34.1	57.6	45.4

🛠️ Quantized Models

GGUF https://huggingface.co/Rebecca19990101/breeze-petro-7b-instruct-v1-q4_k_m.gguf/tree/main

Training Details

Training Data

Dataset: Rebecca19990101/petro-dataset-v2

Downloads last month: 15

Safetensors

Model size

7B params

Tensor type

F32

Model tree for PetroGPT/Breeze-Petro-7B-Instruct-v1

Quantizations

2 models