RichardErkhov/NTQAI_-_Nxcode-CQ-7B-orpo-gguf

Quantization made by Richard Erkhov.

Nxcode-CQ-7B-orpo - GGUF

Model creator: https://huggingface.co/NTQAI/
Original model: https://huggingface.co/NTQAI/Nxcode-CQ-7B-orpo/

Name	Quant method	Size
Nxcode-CQ-7B-orpo.Q2_K.gguf	Q2_K	2.84GB
Nxcode-CQ-7B-orpo.IQ3_XS.gguf	IQ3_XS	3.13GB
Nxcode-CQ-7B-orpo.IQ3_S.gguf	IQ3_S	3.27GB
Nxcode-CQ-7B-orpo.Q3_K_S.gguf	Q3_K_S	3.26GB
Nxcode-CQ-7B-orpo.IQ3_M.gguf	IQ3_M	3.36GB
Nxcode-CQ-7B-orpo.Q3_K.gguf	Q3_K	3.55GB
Nxcode-CQ-7B-orpo.Q3_K_M.gguf	Q3_K_M	3.55GB
Nxcode-CQ-7B-orpo.Q3_K_L.gguf	Q3_K_L	3.71GB
Nxcode-CQ-7B-orpo.IQ4_XS.gguf	IQ4_XS	3.79GB
Nxcode-CQ-7B-orpo.Q4_0.gguf	Q4_0	3.89GB
Nxcode-CQ-7B-orpo.IQ4_NL.gguf	IQ4_NL	3.94GB
Nxcode-CQ-7B-orpo.Q4_K_S.gguf	Q4_K_S	4.11GB
Nxcode-CQ-7B-orpo.Q4_K.gguf	Q4_K	4.41GB
Nxcode-CQ-7B-orpo.Q4_K_M.gguf	Q4_K_M	4.41GB
Nxcode-CQ-7B-orpo.Q4_1.gguf	Q4_1	4.29GB
Nxcode-CQ-7B-orpo.Q5_0.gguf	Q5_0	4.69GB
Nxcode-CQ-7B-orpo.Q5_K_S.gguf	Q5_K_S	4.79GB
Nxcode-CQ-7B-orpo.Q5_K.gguf	Q5_K	5.06GB
Nxcode-CQ-7B-orpo.Q5_K_M.gguf	Q5_K_M	5.06GB
Nxcode-CQ-7B-orpo.Q5_1.gguf	Q5_1	5.09GB
Nxcode-CQ-7B-orpo.Q6_K.gguf	Q6_K	5.94GB
Nxcode-CQ-7B-orpo.Q8_0.gguf	Q8_0	7.18GB

Original model description:

license_name: tongyi-qianwen-research license_link: https://huggingface.co/Qwen/CodeQwen1.5-7B/blob/main/LICENSE tags: - code pipeline_tag: text-generation license: other

Introduction

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.

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.

Quickstart

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)

Contact information

For personal communication related to this project, please contact Nha Nguyen Van ([email protected]).