书生·浦语 官网
HOT
[](https://github.com/open-mmlab/mmdetection/blob/main/LICENSE)
[](https://github.com/internLM/OpenCompass/)
[📘商业授权](#开源许可证) |
[🤗HuggingFace](https://huggingface.co/internlm) |
[🆕最新消息](#更新) |
[🤔提交反馈](https://github.com/InternLM/InternLM/issues/new)|
[📜技术报告](https://arxiv.org/abs/2403.17297)[💬聊天应用](https://internlm-chat.intern-ai.org.cn/) | [🔗API](https://internlm.intern-ai.org.cn/api/document) | [🧩魔乐社区](https://modelers.cn/spaces/MindSpore-Lab/INTERNLM2-20B-PLAN) [English](./README.md) | [简体中文](./README_zh-CN.md)
internlm3_8b_instruct](https://www.modelscope.cn/models/Shanghai_AI_Laboratory/internlm3-8b-instruct/summary) | [](https://modelers.cn/models/Intern/internlm3-8b-instruct) | 2025-01-15 |
### InternLM2.5
#### 深度思考 system prompt
```python
thinking_system_prompt = """You are an expert mathematician with extensive experience in mathematical competitions. You approach problems through systematic thinking and rigorous reasoning. When solving problems, follow these thought processes:
## Deep Understanding
Take time to fully comprehend the problem before attempting a solution. Consider:
- What is the real question being asked?
- What are the given conditions and what do they tell us?
- Are there any special restrictions or assumptions?
- Which information is crucial and which is supplementary?
## Multi-angle Analysis
Before solving, conduct thorough analysis:
- What mathematical concepts and properties are involved?
- Can you recall similar classic problems or solution methods?
- Would diagrams or tables help visualize the problem?
- Are there special cases that need separate consideration?
## Systematic Thinking
Plan your solution path:
- Propose multiple possible approaches
- Analyze the feasibility and merits of each method
- Choose the most appropriate method and explain why
- Break complex problems into smaller, manageable steps
## Rigorous Proof
During the solution process:
- Provide solid justification for each step
- Include detailed proofs for key conclusions
- Pay attention to logical connections
- Be vigilant about potential oversights
## Repeated Verification
After completing your solution:
- Verify your results satisfy all conditions
- Check for overlooked special cases
- Consider if the solution can be optimized or simplified
- Review your reasoning process
Remember:
1. Take time to think thoroughly rather than rushing to an answer
2. Rigorously prove each key conclusion
3. Keep an open mind and try different approaches
4. Summarize valuable problem-solving methods
5. Maintain healthy skepticism and verify multiple times
Your response should reflect deep mathematical understanding and precise logical thinking, making your solution path and reasoning clear to others.
When you're ready, present your complete solution with:
- Clear problem understanding
- Detailed solution process
- Key insights
- Thorough verification
Focus on clear, logical progression of ideas and thorough explanation of your mathematical reasoning. Provide answers in the same language as the user asking the question, repeat the final answer using a '\\boxed{}' without any units, you have [[8192]] tokens to complete the answer.
"""
```
#### Transformers 推理
```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
model_dir = "internlm/internlm3-8b-instruct"
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
model = AutoModelForCausalLM.from_pretrained(model_dir, trust_remote_code=True, torch_dtype=torch.float16)
# (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
# InternLM3 8B in 4bit will cost nearly 8GB GPU memory.
# pip install -U bitsandbytes
# 8-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_8bit=True)
# 4-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_4bit=True)
model = model.eval()
messages = [
{"role": "system", "content": thinking_system_prompt},
{"role": "user", "content": "已知函数\(f(x)=\mathrm{e}^{x}-ax - a^{3}\)。\n（1）当\(a = 1\)时，求曲线\(y = f(x)\)在点\((1,f(1))\)处的切线方程；\n（2）若\(f(x)\)有极小值，且极小值小于\(0\)，求\(a\)的取值范围。"},
]
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
generated_ids = model.generate(tokenized_chat, max_new_tokens=8192)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(tokenized_chat, generated_ids)
]
prompt = tokenizer.batch_decode(tokenized_chat)[0]
print(prompt)
response = tokenizer.batch_decode(generated_ids)[0]
print(response)
```
#### LMDeploy 推理
LMDeploy is a toolkit for compressing, deploying, and serving LLM, developed by the MMRazor and MMDeploy teams.
```bash
pip install lmdeploy
```
You can run batch inference locally with the following python code:
```python
from lmdeploy import pipeline, GenerationConfig, ChatTemplateConfig
model_dir = "internlm/internlm3-8b-instruct"
chat_template_config = ChatTemplateConfig(model_name='internlm3')
pipe = pipeline(model_dir, chat_template_config=chat_template_config)
messages = [
{"role": "system", "content": thinking_system_prompt},
{"role": "user", "content": "已知函数\(f(x)=\mathrm{e}^{x}-ax - a^{3}\)。\n（1）当\(a = 1\)时，求曲线\(y = f(x)\)在点\((1,f(1))\)处的切线方程；\n（2）若\(f(x)\)有极小值，且极小值小于\(0\)，求\(a\)的取值范围。"},
]
response = pipe(messages, gen_config=GenerationConfig(max_new_tokens=2048))
print(response)
```
#### Ollama 推理
安装ollama和拉取模型
```bash
# 安装 ollama
curl -fsSL https://ollama.com/install.sh | sh
# 拉取模型
ollama pull internlm/internlm3-8b-instruct
# 安装python库
pip install ollama
```
推理代码
```python
import ollama
messages = [
{
"role": "system",
"content": thinking_system_prompt,
},
{
"role": "user",
"content": "已知函数\(f(x)=\mathrm{e}^{x}-ax - a^{3}\)。\n（1）当\(a = 1\)时，求曲线\(y = f(x)\)在点\((1,f(1))\)处的切线方程；\n（2）若\(f(x)\)有极小值，且极小值小于\(0\)，求\(a\)的取值范围。"
},
]
stream = ollama.chat(
model='internlm/internlm3-8b-instruct',
messages=messages,
stream=True,
options=dict(num_ctx=8192, num_predict=2048)
)
for chunk in stream:
print(chunk['message']['content'], end='', flush=True)
```
#### vLLM 推理
参考[安装文档](https://docs.vllm.ai/en/latest/getting_started/installation/index.html) 安装 vllm 最新代码
```bash
pip install vllm --pre --extra-index-url https://wheels.vllm.ai/nightly
```
推理代码
```python
from vllm import LLM, SamplingParams
llm = LLM(model="internlm/internlm3-8b-instruct")
sampling_params = SamplingParams(temperature=1, repetition_penalty=1.005, top_k=40, top_p=0.8, max_tokens=8192)
prompts = [
{
"role": "system",
"content": thinking_system_prompt,
},
{
"role": "user",
"content": "已知函数\(f(x)=\mathrm{e}^{x}-ax - a^{3}\)。\n（1）当\(a = 1\)时，求曲线\(y = f(x)\)在点\((1,f(1))\)处的切线方程；\n（2）若\(f(x)\)有极小值，且极小值小于\(0\)，求\(a\)的取值范围。"
},
]
outputs = llm.chat(prompts,
sampling_params=sampling_params,
use_tqdm=False)
print(outputs)
```
## 开源许可证
本仓库的代码和权重依照 Apache-2.0 协议开源。
## 引用
```
@misc{cai2024internlm2,
title={InternLM2 Technical Report},
author={Zheng Cai and Maosong Cao and Haojiong Chen and Kai Chen and Keyu Chen and Xin Chen and Xun Chen and Zehui Chen and Zhi Chen and Pei Chu and Xiaoyi Dong and Haodong Duan and Qi Fan and Zhaoye Fei and Yang Gao and Jiaye Ge and Chenya Gu and Yuzhe Gu and Tao Gui and Aijia Guo and Qipeng Guo and Conghui He and Yingfan Hu and Ting Huang and Tao Jiang and Penglong Jiao and Zhenjiang Jin and Zhikai Lei and Jiaxing Li and Jingwen Li and Linyang Li and Shuaibin Li and Wei Li and Yining Li and Hongwei Liu and Jiangning Liu and Jiawei Hong and Kaiwen Liu and Kuikun Liu and Xiaoran Liu and Chengqi Lv and Haijun Lv and Kai Lv and Li Ma and Runyuan Ma and Zerun Ma and Wenchang Ning and Linke Ouyang and Jiantao Qiu and Yuan Qu and Fukai Shang and Yunfan Shao and Demin Song and Zifan Song and Zhihao Sui and Peng Sun and Yu Sun and Huanze Tang and Bin Wang and Guoteng Wang and Jiaqi Wang and Jiayu Wang and Rui Wang and Yudong Wang and Ziyi Wang and Xingjian Wei and Qizhen Weng and Fan Wu and Yingtong Xiong and Chao Xu and Ruiliang Xu and Hang Yan and Yirong Yan and Xiaogui Yang and Haochen Ye and Huaiyuan Ying and Jia Yu and Jing Yu and Yuhang Zang and Chuyu Zhang and Li Zhang and Pan Zhang and Peng Zhang and Ruijie Zhang and Shuo Zhang and Songyang Zhang and Wenjian Zhang and Wenwei Zhang and Xingcheng Zhang and Xinyue Zhang and Hui Zhao and Qian Zhao and Xiaomeng Zhao and Fengzhe Zhou and Zaida Zhou and Jingming Zhuo and Yicheng Zou and Xipeng Qiu and Yu Qiao and Dahua Lin},
year={2024},
eprint={2403.17297},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```