企业官网建设流程全解析

微信网站怎么做的好名字,开网站平台需要多少钱,北京高端网站建设服务,筑站网络推广一、核心工具 模型微调全流程需安装以下工具#xff1a; 必装工具#xff1a;Unsloth#xff08;高效微调框架#xff09;可选工具#xff1a; vLLM#xff08;模型调度与推理验证#xff09;EvalScope#xff08;模型性能评估#xff09;wandb#xff08;训练过程监…一、核心工具模型微调全流程需安装以下工具必装工具Unsloth高效微调框架可选工具vLLM模型调度与推理验证EvalScope模型性能评估wandb训练过程监控若追求快速上手可先安装 Unsloth若需完整流程验证建议全量安装。二、Unsloth 安装Unsloth 是专为大模型设计的动态量化与微调框架能显著提升微调速度并降低显存占用。虚拟环境创建# 创建虚拟环境 conda create --name unsloth python3.11 conda init source ~/.bashrc conda activate unsloth # 安装Jupyter相关组件 conda install jupyterlab conda install ipykernel python -m ipykernel install --user --name unsloth --display-name Python unsloth框架安装pip install --upgrade --force-reinstall --no-cache-dir unsloth unsloth_zoo安装验证在 Jupyter 中选择 “Python unsloth” 内核运行以下代码测试from unsloth import FastLanguageModel import torch无报错则安装成功。核心优势微调速度提升 2-5 倍显存使用减少 80%兼容 HuggingFace 生态支持 SFT 和 DPO 等微调方式支持 4 位动态量化在低显存设备上也能运行大模型三、vLLM 安装vLLM 是高效的模型推理框架用于微调后模型效果验证。虚拟环境配置# 创建独立虚拟环境AutoDL环境可跳过 conda create --name vllm python3.11 conda init source ~/.bashrc conda activate vllm框架安装# 安装依赖库 pip install bitsandbytes0.45.3 # 安装vLLM pip install --upgrade vllm安装模型下载工具pip install modelscope下载模型权重以 32B 4bit 动态量化模型为例modelscope download --model unsloth/Qwen3-32B-unsloth-bnb-4bit --local_dir ./Qwen3-32B-unsloth-bnb-4bit模型调用测试启动vLLM 服务vllm serve ./Qwen3-32B-unsloth-bnb-4bit --enable-auto-tool-choice --tool-call-parser hermesvLLM调用qwen3参数说明详见https://qwen.readthedocs.io/en/latest/deployment/vllm.html#代码调用测试from openai import OpenAI openai_api_key EMPTY openai_api_base http://localhost:8000/v1 client OpenAI( api_keyopenai_api_key, base_urlopenai_api_base, ) messages [{role: user, content: 你好,好久不见!}] response client.chat.completions.create( model./Qwen3-32B-unsloth-bnb-4bit, messagesmessages, ) print(response.choices[0].message.content)注意Unsloth 动态量化模型仅支持单卡调用32B 4bit 模型最低需 22G 显存调用37G 显存微调目前 vLLM 仅支持 dense 模型MoE 模型暂不兼容四、EvalScope 安装部署流程EvalScope 是大模型性能评估框架用于对比微调前后模型性能变化。4.1 环境搭建# 创建独立虚拟环境 conda create --name evalscope python3.11 conda init source ~/.bashrc conda activate evalscope # 安装Jupyter组件 conda install jupyterlab conda install ipykernel python -m ipykernel install --user --name evalscope --display-name Python evalscope4.2 安装# 基础安装 pip install evalscope # 可选扩展根据需求安装 pip install evalscope[opencompass] # OpenCompass后端 pip install evalscope[vlmeval] # VLMEvalKit后端 pip install evalscope[rag] # RAGEval后端 pip install evalscope[all] # 安装所有后端4.3 压力测试示例evalscope perf \ --url http://127.0.0.1:8000/v1/chat/completions \ --parallel 5 \ --model ./Qwen3-32B-unsloth-bnb-4bit \ --number 20 \ --api openai \ --dataset openqa \ --stream4.4 模型性能评估这里采用了EvalScope专为Qwen3准备的 modelscope/EvalScope-Qwen3-Test 数据集进行评测会围绕模型的推理、指令跟随、代理能力和多语言支持方面能力进行测试该数据包含 mmlu_pro 、ifeval 、 live_code_bench 、 math_500 、 aime24 等各著名评估数据集。数据集地址https://modelscope.cn/datasets/modelscope/EvalScope-Qwen3-Test/summary估代码如下from evalscope import TaskConfig, run_task task_cfg TaskConfig( model./Qwen3-32B-unsloth-bnb-4bit, api_urlhttp://127.0.0.1:8000/v1/chat/completions, eval_typeservice, datasets[data_collection], dataset_args{ data_collection: { dataset_id: modelscope/EvalScope-Qwen3-Test, filters: {remove_until: /think} } }, eval_batch_size128, generation_config{ max_tokens: 30000, temperature: 0.6, top_p: 0.95, top_k: 20, n: 1, }, timeout60000, streamTrue, limit2000, ) run_task(task_cfgtask_cfg)4.5评估结果可视化evalscope app访问输出的本地 URL 即可查看评估报告。五、wandb 安装与注册wandb 是模型训练可视化工具用于监控训练过程中的关键指标。安装pip install wandb注册与登录访问wandb 官网(https://wandb.ai/site)注册账号在终端运行登录命令按提示输入 API 密钥wandb login核心功能实时可视化训练指标损失值、学习率等自动记录实验参数与结果确保可追溯性支持训练中断与恢复多实验对比分析辅助调优决策使用方法在微调代码中加入 wandb 初始化代码import wandb wandb.init(projectqwen3-finetuning, config{epochs: 3, learning_rate: 2e-4})六、相关代码没仔细整理做个记录吧代码来源于网络侵删。from unsloth import FastLanguageModel import torch import requests, json import pandas as pd from datasets import load_dataset from datasets import load_from_disk from unsloth.chat_templates import standardize_sharegpt from trl import SFTTrainer, SFTConfig import wandb import os os.environ[CUDA_VISIBLE_DEVICES] 1 max_seq_length 8192 dtype None load_in_4bit True model, tokenizer FastLanguageModel.from_pretrained( model_name ./Qwen3-32B-unsloth-bnb-4bit, max_seq_length max_seq_length, dtype dtype, load_in_4bit load_in_4bit, ) print(model) print(tokenizer) gpu_stats torch.cuda.get_device_properties(0) start_gpu_memory round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3) max_memory round(gpu_stats.total_memory / 1024 / 1024 / 1024, 3) print(fGPU {gpu_stats.name}. Max memory {max_memory} GB.) print(f{start_gpu_memory} GB of memory reserved.) messages [ {role : user, content : 你好好久不见} ] # 不设置思考 text tokenizer.apply_chat_template( messages, tokenize False, add_generation_prompt True, enable_thinking False, # 设置不思考 ) # |im_start|user\n你好好久不见|im_end|\n|im_start|assistant\nthink\n\n/think\n\n print(text) inputs tokenizer(text, return_tensorspt).to(cuda) outputs model.generate( input_idsinputs.input_ids, attention_maskinputs.attention_mask, max_new_tokensmax_seq_length, use_cacheTrue, ) print(outputs) response tokenizer.batch_decode(outputs) print(response) # 设置思考 text tokenizer.apply_chat_template( messages, tokenize False, add_generation_prompt True, enable_thinking True, # 设置思考 ) inputs tokenizer(text, return_tensorspt).to(cuda) outputs model.generate( input_idsinputs.input_ids, attention_maskinputs.attention_mask, max_new_tokensmax_seq_length, use_cacheTrue, ) response tokenizer.batch_decode(outputs) messages [ {role : system, content : 你是一名助人为乐的助手名叫小明。}, {role : user, content : 你好好久不见请问你叫什么名字} ] text tokenizer.apply_chat_template( messages, tokenize False, add_generation_prompt True, enable_thinking True, # 设置思考 ) inputs tokenizer(text, return_tensorspt).to(cuda) outputs model.generate( input_idsinputs.input_ids, attention_maskinputs.attention_mask, max_new_tokensmax_seq_length, use_cacheTrue, ) response tokenizer.batch_decode(outputs) # function calling def get_weather(loc): 查询即时天气函数 :param loc: 必要参数字符串类型用于表示查询天气的具体城市名称\ 注意中国的城市需要用对应城市的英文名称代替例如如果需要查询北京市天气则loc参数需要输入Beijing :returnOpenWeather API查询即时天气的结果具体URL请求地址为https://api.openweathermap.org/data/2.5/weather\ 返回结果对象类型为解析之后的JSON格式对象并用字符串形式进行表示其中包含了全部重要的天气信息 # Step 1.构建请求 url https://api.openweathermap.org/data/2.5/weather # Step 2.设置查询参数 params { q: loc, appid: YOUR_API_KEY, # 输入API key units: metric, # 使用摄氏度而不是华氏度 lang:zh_cn # 输出语言为简体中文 } # Step 3.发送GET请求 response requests.get(url, paramsparams) # Step 4.解析响应 data response.json() return json.dumps(data) tools [ { type: function, function:{ name: get_weather, description: 查询即时天气函数根据输入的城市名称查询对应城市的实时天气一次只能输入一个城市名称, parameters: { type: object, properties: { loc: { description: 城市名称注意中国的城市需要用对应城市的英文名称代替例如如果需要查询北京市天气则loc参数需要输入Beijing, type: string } }, required: [loc] } } } ] messages [ {role : system, content : 你是一名助人为乐的天气查询助手当用户询问天气信息时请调用get_weather函数进行天气查询。}, {role : user, content : 你好请帮我查询下北京今天天气如何} ] text tokenizer.apply_chat_template( messages, tools tools, tokenize False, add_generation_prompt True, enable_thinking True, # 设置思考 ) inputs tokenizer(text, return_tensorspt).to(cuda) outputs model.generate( input_idsinputs.input_ids, attention_maskinputs.attention_mask, max_new_tokensmax_seq_length, use_cacheTrue, ) response tokenizer.batch_decode(outputs) messages [ {role : system, content : 你是一名助人为乐的天气查询助手当用户询问天气信息时请调用get_weather函数进行天气查询。}, {role : user, content : 你好请帮我查询下北京和杭州今天天气如何} ] text tokenizer.apply_chat_template( messages, tools tools, tokenize False, add_generation_prompt True, enable_thinking True, # 设置思考 ) inputs tokenizer(text, return_tensorspt).to(cuda) outputs model.generate( input_idsinputs.input_ids, attention_maskinputs.attention_mask, max_new_tokensmax_seq_length, use_cacheTrue, ) response tokenizer.batch_decode(outputs) messages.append({ role: assistant, content: think\n我将调用 get_weather 函数来查询天气。\n/think\n, tool_calls: [ { name: get_weather, arguments: { location: 北京 } }, { name: get_weather, arguments: { location: 杭州 } } ] }) messages.append({ role: tool, content: json.dumps({ location: 北京, weather: 晴最高气温26℃ }) }) messages.append({ role: tool, content: json.dumps({ location: 杭州, weather: 多云转小雨最高气温23℃ }) }) text tokenizer.apply_chat_template( messages, tools tools, tokenize False, add_generation_prompt True, enable_thinking True, # 设置思考 ) inputs tokenizer(text, return_tensorspt).to(cuda) outputs model.generate( input_idsinputs.input_ids, attention_maskinputs.attention_mask, max_new_tokensmax_seq_length, use_cacheTrue, ) response tokenizer.batch_decode(outputs) # stream 模式 from transformers import TextStreamer messages [ {role : user, content : 你好好久不见} ] text tokenizer.apply_chat_template( messages, tokenize False, add_generation_prompt True, enable_thinking False, ) _ model.generate( **tokenizer(text, return_tensors pt).to(cuda), max_new_tokens 256, # Increase for longer outputs! temperature 0.7, top_p 0.8, top_k 20, # For non thinking streamer TextStreamer(tokenizer, skip_prompt True), ) text tokenizer.apply_chat_template( messages, tokenize False, add_generation_prompt True, enable_thinking True, ) _ model.generate( **tokenizer(text, return_tensors pt).to(cuda), max_new_tokens 2048, # Increase for longer outputs! temperature 0.6, top_p 0.95, top_k 20, # For thinking streamer TextStreamer(tokenizer, skip_prompt True), ) reasoning_dataset load_dataset(unsloth/OpenMathReasoning-mini, split cot) non_reasoning_dataset load_dataset(mlabonne/FineTome-100k, split train) print(len(reasoning_dataset), reasoning_dataset[0]) def generate_conversation(examples): problems examples[problem] solutions examples[generated_solution] conversations [] for problem, solution in zip(problems, solutions): conversations.append([ {role : user, content : problem}, {role : assistant, content : solution}, ]) return { conversations: conversations, } reasoning_data reasoning_dataset.map(generate_conversation, batched True) print(reasoning_data[conversations][0]) reasoning_conversations tokenizer.apply_chat_template( reasoning_data[conversations], tokenize False, ) print(len(reasoning_conversations), reasoning_conversations[0]) dataset standardize_sharegpt(non_reasoning_dataset) print(dataset[conversations][0]) non_reasoning_conversations tokenizer.apply_chat_template( dataset[conversations], tokenize False, ) print(non_reasoning_conversations[0]) print(len(reasoning_conversations)) print(len(non_reasoning_conversations)) chat_percentage 0.75 non_reasoning_subset pd.Series(non_reasoning_conversations) non_reasoning_subset non_reasoning_subset.sample( int(len(reasoning_conversations) * (1.0 - chat_percentage)), random_state 2407, ) data pd.concat([ pd.Series(reasoning_conversations), pd.Series(non_reasoning_subset) ]) data.name text from datasets import Dataset combined_dataset Dataset.from_pandas(pd.DataFrame(data)) combined_dataset combined_dataset.shuffle(seed 3407) print(len(combined_dataset), type(combined_dataset)) combined_dataset.save_to_disk(cleaned_qwen3_dataset) combined_dataset load_from_disk(cleaned_qwen3_dataset) print(type(combined_dataset)) model FastLanguageModel.get_peft_model( model, r 32, # Choose any number 0! Suggested 8, 16, 32, 64, 128 target_modules [q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj,], lora_alpha 32, # Best to choose alpha rank or rank*2 lora_dropout 0, # Supports any, but 0 is optimized bias none, # Supports any, but none is optimized # [NEW] unsloth uses 30% less VRAM, fits 2x larger batch sizes! use_gradient_checkpointing unsloth, # True or unsloth for very long context random_state 3407, use_rslora False, # We support rank stabilized LoRA loftq_config None, # And LoftQ ) trainer SFTTrainer( model model, tokenizer tokenizer, train_dataset combined_dataset, eval_dataset None, # Can set up evaluation! args SFTConfig( dataset_text_field text, per_device_train_batch_size 4, gradient_accumulation_steps 2, # Use GA to mimic batch size! warmup_steps 5, num_train_epochs 1, # Set this for 1 full training run. learning_rate 2e-4, # Reduce to 2e-5 for long training runs logging_steps 1, optim adamw_8bit, weight_decay 0.01, lr_scheduler_type linear, seed 3407, report_to none, # Use this for WandB etc ), ) # title Show current memory stats gpu_stats torch.cuda.get_device_properties(0) start_gpu_memory round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3) max_memory round(gpu_stats.total_memory / 1024 / 1024 / 1024, 3) print(fGPU {gpu_stats.name}. Max memory {max_memory} GB.) print(f{start_gpu_memory} GB of memory reserved.) os.environ[WANDB_NOTEBOOK_NAME] notebook_name.ipynb wandb.login(keyyour_key) run wandb.init(projectFine-tune-Qwen-32B-4bit on Combined Dataset, ) trainer_stats trainer.train() # title Show final memory and time stats used_memory round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3) used_memory_for_lora round(used_memory - start_gpu_memory, 3) used_percentage round(used_memory / max_memory * 100, 3) lora_percentage round(used_memory_for_lora / max_memory * 100, 3) print(f{trainer_stats.metrics[train_runtime]} seconds used for training.) print( f{round(trainer_stats.metrics[train_runtime]/60, 2)} minutes used for training. ) print(fPeak reserved memory {used_memory} GB.) print(fPeak reserved memory for training {used_memory_for_lora} GB.) print(fPeak reserved memory % of max memory {used_percentage} %.) print(fPeak reserved memory for training % of max memory {lora_percentage} %.) print(model) messages [ {role : user, content : Determine the surface area of the portion of the plane $2x 3y 6z 9$ that lies in the first octant.} ] text tokenizer.apply_chat_template( messages, tokenize False, add_generation_prompt True, # Must add for generation enable_thinking True, # Disable thinking ) _ model.generate( **tokenizer(text, return_tensors pt).to(cuda), max_new_tokens 20488, # Increase for longer outputs! temperature 0.6, top_p 0.95, top_k 20, # For thinking streamer TextStreamer(tokenizer, skip_prompt True), ) model.save_pretrained_merged(save_directory Qwen3-32B-finetuned-fp16, tokenizer tokenizer, save_method merged_16bit) max_seq_length 8192 dtype None load_in_4bit True model, tokenizer FastLanguageModel.from_pretrained( model_name ./Qwen3-8B-unsloth-bnb-4bit, max_seq_length max_seq_length, dtype dtype, load_in_4bit load_in_4bit, ) question_1 中国在哪里 messages [ {role : user, content : question_1} ] text tokenizer.apply_chat_template( messages, tokenize False, add_generation_prompt True, # Must add for generation enable_thinking True, # Disable thinking ) _ model.generate( **tokenizer(text, return_tensors pt).to(cuda), max_new_tokens 20488, # Increase for longer outputs! temperature 0.6, top_p 0.95, top_k 20, # For thinking streamer TextStreamer(tokenizer, skip_prompt True), ) def load_jsonl_dataset(file_path): 从JSONL文件加载数据集 data {input: [], output: []} print(f开始加载数据集: {file_path}) count 0 error_count 0 with open(file_path, r, encodingutf-8) as f: for line in f: try: item json.loads(line.strip()) # 根据数据集结构提取字段 input_text item.get(input, ) output item.get(output, ) data[input].append(input_text) data[output].append(output) count 1 except Exception as e: print(f解析行时出错: {e}) error_count 1 continue print(f数据集加载完成: 成功加载{count}个样本, 跳过{error_count}个错误样本) return Dataset.from_dict(data) data_path ./train_1k.jsonl # 加载自定义数据集 dataset load_jsonl_dataset(data_path) # 显示数据集信息 print(f\n数据集统计:) print(f- 样本数量: {len(dataset)}) print(f- 字段: {dataset.column_names}) print(dataset[0]) def formatting_prompts_func(examples): 根据提示模板格式化数据 inputs examples[input] outputs examples[output] texts [] for input_text, output in zip(inputs, outputs): texts.append([ {role : user, content : input_text}, {role : assistant, content : output}, ]) return {text: texts} # 应用格式化 print(开始格式化数据集...) reasoning_conversations tokenizer.apply_chat_template( dataset.map(formatting_prompts_func, batched True)[text], tokenize False, ) print(数据集格式化完成) non_reasoning_dataset load_dataset(mlabonne/FineTome-100k, split train) dataset standardize_sharegpt(non_reasoning_dataset) non_reasoning_conversations tokenizer.apply_chat_template( dataset[conversations], tokenize False, ) non_reasoning_subset pd.Series(non_reasoning_conversations) non_reasoning_subset non_reasoning_subset.sample( 1000, random_state 2407, ) data pd.concat([ pd.Series(reasoning_conversations), pd.Series(non_reasoning_subset) ]) data.name text combined_dataset Dataset.from_pandas(pd.DataFrame(data)) combined_dataset combined_dataset.shuffle(seed 3407) model FastLanguageModel.get_peft_model( model, r 32, # Choose any number 0! Suggested 8, 16, 32, 64, 128 target_modules [q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj,], lora_alpha 32, # Best to choose alpha rank or rank*2 lora_dropout 0, # Supports any, but 0 is optimized bias none, # Supports any, but none is optimized # [NEW] unsloth uses 30% less VRAM, fits 2x larger batch sizes! use_gradient_checkpointing unsloth, # True or unsloth for very long context random_state 3407, use_rslora False, # We support rank stabilized LoRA loftq_config None, # And LoftQ ) trainer SFTTrainer( model model, tokenizer tokenizer, train_dataset combined_dataset, eval_dataset None, # Can set up evaluation! args SFTConfig( dataset_text_field text, per_device_train_batch_size 2, gradient_accumulation_steps 4, # Use GA to mimic batch size! warmup_steps 5, num_train_epochs 1, # Set this for 1 full training run. learning_rate 2e-4, # Reduce to 2e-5 for long training runs logging_steps 1, optim adamw_8bit, weight_decay 0.01, lr_scheduler_type linear, seed 3407, report_to None, # Use this for WandB etc ), ) # title Show current memory stats gpu_stats torch.cuda.get_device_properties(0) start_gpu_memory round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3) max_memory round(gpu_stats.total_memory / 1024 / 1024 / 1024, 3) print(fGPU {gpu_stats.name}. Max memory {max_memory} GB.) print(f{start_gpu_memory} GB of memory reserved.) trainer_stats trainer.train() # title Show final memory and time stats used_memory round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3) used_memory_for_lora round(used_memory - start_gpu_memory, 3) used_percentage round(used_memory / max_memory * 100, 3) lora_percentage round(used_memory_for_lora / max_memory * 100, 3) print(f{trainer_stats.metrics[train_runtime]} seconds used for training.) print( f{round(trainer_stats.metrics[train_runtime]/60, 2)} minutes used for training. ) print(fPeak reserved memory {used_memory} GB.) print(fPeak reserved memory for training {used_memory_for_lora} GB.) print(fPeak reserved memory % of max memory {used_percentage} %.) print(fPeak reserved memory for training % of max memory {lora_percentage} %.) messages [ {role : user, content : question_1} ] text tokenizer.apply_chat_template( messages, tokenize False, add_generation_prompt True, # Must add for generation enable_thinking True, # Disable thinking ) _ model.generate( **tokenizer(text, return_tensors pt).to(cuda), max_new_tokens 20488, # Increase for longer outputs! temperature 0.6, top_p 0.95, top_k 20, # For thinking streamer TextStreamer(tokenizer, skip_prompt True), )想入门 AI 大模型却找不到清晰方向备考大厂 AI 岗还在四处搜集零散资料别再浪费时间啦2025 年AI 大模型全套学习资料已整理完毕从学习路线到面试真题从工具教程到行业报告一站式覆盖你的所有需求现在全部免费分享扫码免费领取全部内容​一、学习必备100本大模型电子书26 份行业报告 600 套技术PPT帮你看透 AI 趋势想了解大模型的行业动态、商业落地案例大模型电子书这份资料帮你站在 “行业高度” 学 AI1. 100本大模型方向电子书2. 26 份行业研究报告覆盖多领域实践与趋势报告包含阿里、DeepSeek 等权威机构发布的核心内容涵盖职业趋势《AI 职业趋势报告》《中国 AI 人才粮仓模型解析》商业落地《生成式 AI 商业落地白皮书》《AI Agent 应用落地技术白皮书》领域细分《AGI 在金融领域的应用报告》《AI GC 实践案例集》行业监测《2024 年中国大模型季度监测报告》《2025 年中国技术市场发展趋势》。3. 600套技术大会 PPT听行业大咖讲实战PPT 整理自 2024-2025 年热门技术大会包含百度、腾讯、字节等企业的一线实践安全方向《端侧大模型的安全建设》《大模型驱动安全升级腾讯代码安全实践》产品与创新《大模型产品如何创新与创收》《AI 时代的新范式构建 AI 产品》多模态与 Agent《Step-Video 开源模型视频生成进展》《Agentic RAG 的现在与未来》工程落地《从原型到生产AgentOps 加速字节 AI 应用落地》《智能代码助手 CodeFuse 的架构设计》。二、求职必看大厂 AI 岗面试 “弹药库”300 真题 107 道面经直接抱走想冲字节、腾讯、阿里、蔚来等大厂 AI 岗这份面试资料帮你提前 “押题”拒绝临场慌1. 107 道大厂面经覆盖 Prompt、RAG、大模型应用工程师等热门岗位面经整理自 2021-2025 年真实面试场景包含 TPlink、字节、腾讯、蔚来、虾皮、中兴、科大讯飞、京东等企业的高频考题每道题都附带思路解析2. 102 道 AI 大模型真题直击大模型核心考点针对大模型专属考题从概念到实践全面覆盖帮你理清底层逻辑3. 97 道 LLMs 真题聚焦大型语言模型高频问题专门拆解 LLMs 的核心痛点与解决方案比如让很多人头疼的 “复读机问题”三、路线必明 AI 大模型学习路线图1 张图理清核心内容刚接触 AI 大模型不知道该从哪学起这份「AI大模型 学习路线图」直接帮你划重点不用再盲目摸索路线图涵盖 5 大核心板块从基础到进阶层层递进一步步带你从入门到进阶从理论到实战。L1阶段:启航篇丨极速破界AI新时代L1阶段了解大模型的基础知识以及大模型在各个行业的应用和分析学习理解大模型的核心原理、关键技术以及大模型应用场景。L2阶段攻坚篇丨RAG开发实战工坊L2阶段AI大模型RAG应用开发工程主要学习RAG检索增强生成包括Naive RAG、Advanced-RAG以及RAG性能评估还有GraphRAG在内的多个RAG热门项目的分析。L3阶段跃迁篇丨Agent智能体架构设计L3阶段大模型Agent应用架构进阶实现主要学习LangChain、 LIamaIndex框架也会学习到AutoGPT、 MetaGPT等多Agent系统打造Agent智能体。L4阶段精进篇丨模型微调与私有化部署L4阶段大模型的微调和私有化部署更加深入的探讨Transformer架构学习大模型的微调技术利用DeepSpeed、Lamam Factory等工具快速进行模型微调并通过Ollama、vLLM等推理部署框架实现模型的快速部署。L5阶段专题集丨特训篇 【录播课】四、资料领取全套内容免费抱走学 AI 不用再找第二份不管你是 0 基础想入门 AI 大模型还是有基础想冲刺大厂、了解行业趋势这份资料都能满足你现在只需按照提示操作就能免费领取扫码免费领取全部内容​2025 年想抓住 AI 大模型的风口别犹豫这份免费资料就是你的 “起跑线”