MiniCPM3-4B-GPTQ-Int4

Image: image

MiniCPM Repo | MiniCPM Paper | MiniCPM-V Repo | Join us in Discord and WeChat

Introduction#

MiniCPM3-4B is the 3rd generation of MiniCPM series. The overall performance of MiniCPM3-4B surpasses Phi-3.5-mini-Instruct and GPT-3.5-Turbo-0125, being comparable with many recent 7B~9B models.

Compared to MiniCPM1.0/MiniCPM2.0, MiniCPM3-4B has a more powerful and versatile skill set to enable more general usage. MiniCPM3-4B supports function call, along with code interpreter. Please refer to Advanced Features for usage guidelines.

MiniCPM3-4B has a 32k context window. Equipped with LLMxMapReduce, MiniCPM3-4B can handle infinite context theoretically, without requiring huge amount of memory.

Usage#

Inference with Transformers#

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

path = "openbmb/MiniCPM3-4B-GPTQ-Int4"
device = "cuda"

tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(path, torch_dtype=torch.float16, device_map=device, trust_remote_code=True)

messages = [
    {"role": "user", "content": "推荐5个北京的景点。"},
]
model_inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True).to(device)

model_outputs = model.generate(
    model_inputs,
    max_new_tokens=1024,
    top_p=0.7,
    temperature=0.7
)

output_token_ids = [
    model_outputs[i][len(model_inputs[i]):] for i in range(len(model_inputs))
]

responses = tokenizer.batch_decode(output_token_ids, skip_special_tokens=True)[0]
print(responses)

Inference with vLLM#

For now, you need to install our forked version of vLLM.

pip install git+https://github.com/OpenBMB/vllm.git@minicpm3

from transformers import AutoTokenizer
from vllm import LLM, SamplingParams

model_name = "openbmb/MiniCPM3-4B-GPTQ-Int4"
prompt = [{"role": "user", "content": "推荐5个北京的景点。"}]

tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
input_text = tokenizer.apply_chat_template(prompt, tokenize=False, add_generation_prompt=True)

llm = LLM(
    model=model_name,
    trust_remote_code=True,
    tensor_parallel_size=1,
    quantization='gptq'
)
sampling_params = SamplingParams(top_p=0.7, temperature=0.7, max_tokens=1024, repetition_penalty=1.02)

outputs = llm.generate(prompts=input_text, sampling_params=sampling_params)

print(outputs[0].outputs[0].text)

Evaluation Results#

| Benchmark | Qwen2-7B-Instruct | GLM-4-9B-Chat | Gemma2-9B-it | Llama3.1-8B-Instruct | GPT-3.5-Turbo-0125 | Phi-3.5-mini-Instruct(3.8B) | MiniCPM3-4B

| English

| MMLU | 70.5 | 72.4 | 72.6 | 69.4 | 69.2 | 68.4 | 67.2

| BBH | 64.9 | 76.3 | 65.2 | 67.8 | 70.3 | 68.6 | 70.2

| MT-Bench | 8.41 | 8.35 | 7.88 | 8.28 | 8.17 | 8.60 | 8.41

| IFEVAL (Prompt Strict-Acc.) | 51.0 | 64.5 | 71.9 | 71.5 | 58.8 | 49.4 | 68.4

| Chinese

| CMMLU | 80.9 | 71.5 | 59.5 | 55.8 | 54.5 | 46.9 | 73.3

| CEVAL | 77.2 | 75.6 | 56.7 | 55.2 | 52.8 | 46.1 | 73.6

| AlignBench v1.1 | 7.10 | 6.61 | 7.10 | 5.68 | 5.82 | 5.73 | 6.74

| FollowBench-zh (SSR) | 63.0 | 56.4 | 57.0 | 50.6 | 64.6 | 58.1 | 66.8

| Math

| MATH | 49.6 | 50.6 | 46.0 | 51.9 | 41.8 | 46.4 | 46.6

| GSM8K | 82.3 | 79.6 | 79.7 | 84.5 | 76.4 | 82.7 | 81.1

| MathBench | 63.4 | 59.4 | 45.8 | 54.3 | 48.9 | 54.9 | 65.6

| Code

| HumanEval+ | 70.1 | 67.1 | 61.6 | 62.8 | 66.5 | 68.9 | 68.3

| MBPP+ | 57.1 | 62.2 | 64.3 | 55.3 | 71.4 | 55.8 | 63.2

| LiveCodeBench v3 | 22.2 | 20.2 | 19.2 | 20.4 | 24.0 | 19.6 | 22.6

| Function Call

| BFCL v2 | 71.6 | 70.1 | 19.2 | 73.3 | 75.4 | 48.4 | 76.0

| Overall

| Average | 65.3 | 65.0 | 57.9 | 60.8 | 61.0 | 57.2 | 66.3

Statement#

• As a language model, MiniCPM3-4B generates content by learning from a vast amount of text.
• However, it does not possess the ability to comprehend or express personal opinions or value judgments.
• Any content generated by MiniCPM3-4B does not represent the viewpoints or positions of the model developers.
• Therefore, when using content generated by MiniCPM3-4B, users should take full responsibility for evaluating and verifying it on their own.

LICENSE#

• This repository is released under the Apache-2.0 License.
• The usage of MiniCPM3-4B model weights must strictly follow MiniCPM Model License.md.
• The models and weights of MiniCPM3-4B are completely free for academic research. after filling out a "questionnaire" for registration, are also available for free commercial use.

Citation#

@article{hu2024minicpm,
  title={MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies},
  author={Hu, Shengding and Tu, Yuge and Han, Xu and He, Chaoqun and Cui, Ganqu and Long, Xiang and Zheng, Zhi and Fang, Yewei and Huang, Yuxiang and Zhao, Weilin and others},
  journal={arXiv preprint arXiv:2404.06395},
  year={2024}
}