Nxcode-CQ-7B-orpo Open-Source Code Generation Model - Achieve Efficient Code Output through Fine-Tuning on High-Quality Data

Nxcode CQ 7B Orpo

Developed by NTQAI

A code generation model fine-tuned on 100,000 high-quality ranked data points using reference-free monolithic preference optimization, based on the Qwen/CodeQwen1.5-7B model

Large Language Model

Transformers

OtherOpen Source License:Other #Code Generation Optimization #Reference-Free ORPO #Python Completion

Downloads 3,387

Release Time : 4/24/2024

Model Overview

Nxcode-CQ-7B-orpo is an AI model focused on code generation, improving code quality and accuracy through optimization methods

Model Features

High-Quality Code Generation

Fine-tuned on 100,000 high-quality ranked data points, generating superior code quality

Advanced Optimization Method

Utilizes reference-free monolithic preference optimization to enhance model performance

Outstanding Evaluation Performance

Demonstrates excellent results on benchmarks like HumanEval and MBPP

Model Capabilities

Python code completion

Function generation

Code comprehension

Programming problem solving

Use Cases

Programming Assistance

Function Completion

Automatically completes full function implementations based on function signatures and descriptions

Achieves 86.6 pass@1 score on HumanEval

Programming Exercise Solutions

Helps solve programming exercises and algorithm problems

Achieves 82.3 pass@1 score on MBPP

Development Tools

Code Snippet Generation

Generates usable code snippets from natural language descriptions

🚀 Nxcode-CQ-7B-orpo

Nxcode-CQ-7B-orpo is a fine - tuned model based on Qwen/CodeQwen1.5 - 7B, optimized for code generation tasks.

🚀 Quick Start

Here provides a code snippet with apply_chat_template to show you how to load the tokenizer and model and how to generate contents. You should upgrade the transformers if you receive an error when loading the tokenizer

from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # the device to load the model onto

model = AutoModelForCausalLM.from_pretrained(
    "NTQAI/Nxcode-CQ-7B-orpo",
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("NTQAI/Nxcode-CQ-7B-orpo")

prompt = """Complete the following Python function:
from typing import List


def has_close_elements(numbers: List[float], threshold: float) -> bool:
    """ Check if in given list of numbers, are any two numbers closer to each other than
    given threshold.
    >>> has_close_elements([1.0, 2.0, 3.0], 0.5)
    False
    >>> has_close_elements([1.0, 2.8, 3.0, 4.0, 5.0, 2.0], 0.3)
    True
    """
"""
messages = [
    {"role": "user", "content": prompt}
]

inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
outputs = model.generate(inputs, max_new_tokens=512, do_sample=False, top_k=50, top_p=0.95, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id)
res = tokenizer.decode(outputs[0][len(inputs[0]):], skip_special_tokens=True)

✨ Features

Nxcode-CQ-7B-orpo is an Monolithic Preference Optimization without Reference Model fine - tune of Qwen/CodeQwen1.5-7B on 100k samples of high - quality ranking data.

📚 Documentation

Evalplus

EvalPlus	pass@1
HumanEval	86.6
HumanEval+	83.5
MBPP(v0.2.0)	82.3
MBPP+(v0.2.0)	70.4

We use a simple template to generate the solution for evalplus:

"Complete the following Python function:\n{prompt}"

Evalplus Leaderboard

Models	HumanEval	HumanEval+
GPT-4-Turbo (April 2024)	90.2	86.6
GPT-4 (May 2023)	88.4	81.17
GPT-4-Turbo (Nov 2023)	85.4	79.3
CodeQwen1.5-7B-Chat	83.5	78.7
claude-3-opus (Mar 2024)	82.9	76.8
DeepSeek-Coder-33B-instruct	81.1	75.0
WizardCoder-33B-V1.1	79.9	73.2
OpenCodeInterpreter-DS-33B	79.3	73.8
speechless-codellama-34B-v2.0	77.4	72
GPT-3.5-Turbo (Nov 2023)	76.8	70.7
Llama3-70B-instruct	76.2	70.7

Bigcode Leaderboard

09/05/2024

Top 1 average score.

Top 2 winrate.

image/png

📄 License

This project is under the tongyi-qianwen-research license.

Contact information

For personal communication related to this project, please contact Nha Nguyen Van (nha.nguyen@ntq-solution.com.vn).

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご