3 lines
38 KiB
Java
3 lines
38 KiB
Java
var de=Object.defineProperty;var Y=Object.getOwnPropertySymbols;var ue=Object.prototype.hasOwnProperty,ce=Object.prototype.propertyIsEnumerable;var j=(l,e,s)=>e in l?de(l,e,{enumerable:!0,configurable:!0,writable:!0,value:s}):l[e]=s,$=(l,e)=>{for(var s in e||(e={}))ue.call(e,s)&&j(l,s,e[s]);if(Y)for(var s of Y(e))ce.call(e,s)&&j(l,s,e[s]);return l};var R=(l,e,s)=>new Promise((a,k)=>{var P=v=>{try{d(s.next(v))}catch(h){k(h)}},B=v=>{try{d(s.throw(v))}catch(h){k(h)}},d=v=>v.done?a(v.value):Promise.resolve(v.value).then(P,B);d((s=s.apply(l,e)).next())});import{f as m,ah as g,aC as u,as as i,aE as pe,aF as n,au as c,ar as M,ag as p,at as me,G as y,k as b,av as H,F as X,aD as J,A as ee}from"./vue-vendor-C7Zq48Yl.js";import{M as ge}from"./BasicModal-0sAdFEk_.js";import"./index-Du0A3ksf.js";import{f as N,u as Ae,ak as be,aO as ve,d as ye}from"./index-BI6CMai0.js";import{M as fe,aN as Ee,h as Be}from"./antd-vue-vendor-BPnV8VqP.js";import{B as he}from"./BasicForm-BJnSRkIE.js";import"./index-Di-j2gt4.js";import Me from"./AiModelSeniorForm-D-N7zKNj.js";import{u as xe}from"./useForm-BZtHZyx2.js";const we=JSON.parse('[{"title":"Anthropic","value":"ANTHROPIC","LLM":[{"label":"claude-sonnet-4-20250514","value":"claude-sonnet-4-20250514","descr":"【Claude 4系列】Claude Sonnet 4具有卓越推理能力的高性能模型。\\n\\n支持文本和图像输入，文本输出，拥有200k上下文窗口（1M上下文测试版可用）。","type":"text,image"},{"label":"claude-opus-4-20250514","value":"claude-opus-4-20250514","descr":"【Claude 4系列】Claude Opus 4是最强大、最有能力的模型。\\n\\n支持文本和图像输入，文本输出，拥有200k上下文窗口，卓越的推理能力。","type":"text,image"},{"label":"claude-3-7-sonnet-20250219","value":"claude-3-7-sonnet-20250219","descr":"【Claude 3.7系列】Claude Sonnet 3.7中型模型，具备卓越的推理能力和效率。\\n\\n适合企业工作负载和大规模AI部署。","type":"text,image"},{"label":"claude-3-5-sonnet-20241022","value":"claude-3-5-sonnet-20241022","descr":"Claude 3.5 Sonnet是Anthropic最强大的AI模型。\\n\\n在编程、多步骤工作流、图表解释等复杂任务中表现出色。支持200k上下文长度，支持8k最大输出。","type":"text,image"},{"label":"claude-3-5-haiku-20241022","value":"claude-3-5-haiku-20241022","descr":"【快速模型】Claude 3.5 Haiku是Anthropic最快的AI模型。\\n\\n响应速度快，成本较低，适合高频交互场景。支持200k上下文长度，支持8k最大输出。","type":"text,image"},{"label":"claude-3-opus-20240229","value":"claude-3-opus-20240229","descr":"Claude 3 Opus是Claude 3系列中性能最强的模型。\\n\\n在高度复杂的任务上表现出色，如编写高质量代码、数学推理等。支持200k上下文长度，支持4k最大输出。","type":"text,image"},{"label":"claude-3-sonnet-20240229","value":"claude-3-sonnet-20240229","descr":"Claude 3 Sonnet在智能和速度之间取得了良好平衡。\\n\\n适合企业工作负载和大规模AI部署。支持200k上下文长度，支持4k最大输出。","type":"text,image"},{"label":"claude-3-haiku-20240307","value":"claude-3-haiku-20240307","descr":"Claude 3 Haiku是Claude 3系列中最快的模型。\\n\\n提供接近即时的响应，适合无缝AI体验。支持200k上下文长度，支持4k最大输出。","type":"text,image"}],"type":["LLM"],"baseUrl":"https://api.anthropic.com/v1","LLMDefaultValue":"claude-3-5-sonnet-20241022"},{"title":"DeepSeek","value":"DEEPSEEK","LLM":[{"label":"deepseek-reasoner","value":"deepseek-reasoner","descr":"【官方模型】深度求索 新推出的推理模型R1满血版\\n火便全球。\\n支持64k上下文，其中支持8k最大回复。","type":"text"},{"label":"deepseek-chat","value":"deepseek-chat","descr":"最强开源 MoE 模型 DeepSeek-V3，全球首个在代码、数学能力上与GPT-4-Turbo争锋的模型，在代码、数学的多个榜单上位居全球第二；","type":"text"}],"type":["LLM"],"baseUrl":"https://api.deepseek.com/v1","LLMDefaultValue":"deepseek-chat"},{"title":"Ollama","value":"OLLAMA","LLM":[{"label":"llama2","value":"llama2"},{"label":"llama2:13b","value":"llama2:13b"},{"label":"llama2:70b","value":"llama2:70b"},{"label":"llama2-chinese:13b","value":"llama2-chinese:13b"},{"label":"llama3:8b","value":"llama3:8b"},{"label":"llama3:70b","value":"llama3:70b"},{"label":"qwen:0.5b","value":"qwen:0.5b"},{"label":"qwen:1.8b","value":"qwen:1.8b"},{"label":"qwen:4b","value":"qwen:4b"},{"label":"qwen:7b","value":"qwen:7b"},{"label":"qwen:14b","value":"qwen:14b"},{"label":"qwen:32b","value":"qwen:32b"},{"label":"qwen:72b","value":"qwen:72b"},{"label":"qwen:110b","value":"qwen:110b"},{"label":"qwen2:72b-instruct","value":"qwen2:72b-instruct"},{"label":"qwen2:57b-a14b-instruct","value":"qwen2:57b-a14b-instruct"},{"label":"qwen2:7b-instruct","value":"qwen2:7b-instruct"},{"label":"qwen2.5:72b-instruct","value":"qwen2.5:72b-instruct"},{"label":"qwen2.5:32b-instruct","value":"qwen2.5:32b-instruct"},{"label":"qwen2.5:14b-instruct","value":"qwen2.5:14b-instruct"},{"label":"qwen2.5:7b-instruct","value":"qwen2.5:7b-instruct"},{"label":"qwen2.5:1.5b-instruct","value":"qwen2.5:1.5b-instruct"},{"label":"qwen2.5:0.5b-instruct","value":"qwen2.5:0.5b-instruct"},{"label":"qwen2.5:3b-instruct","value":"qwen2.5:3b-instruct"},{"label":"phi3","value":"phi3"}],"EMBED":[{"label":"nomic-embed-text","value":"nomic-embed-text"}],"type":["LLM","EMBED"],"baseUrl":"http://localhost:11434","LLMDefaultValue":"llama2","EMBEDDefaultValue":"nomic-embed-text"},{"title":"OpenAI","value":"OPENAI","LLM":[{"label":"gpt-3.5-turbo","value":"gpt-3.5-turbo","descr":"纯官方高速GPT3.5系列，目前指向gpt-35-turbo-0125模型，最大回复小于4k。\\n综合能力强，过去使用最广泛的文本模型。","type":"text"},{"label":"gpt-4","value":"gpt-4","descr":"纯官方GPT4系列。知识库截止于2021年，价格适中，具有中等参数，比gpt-4turbo系列略强。","type":"text"},{"label":"gpt-4o","value":"gpt-4o","descr":"GPT-4o，是openai的新旗舰型号，支持文本和图片分析。\\n\\n是迈向更自然的人机交互的一步——它接受文本和图像的任意组合作为输入，并生成文本和图像输出的任意组合。","type":"text,image"},{"label":"gpt-4o-mini","value":"gpt-4o-mini","descr":"GPT-4o mini是目前性价比最高的小参数模型，性能介于GPT3.5~GPT4o之间。\\n\\n成本相比GPT-3.5 Turbo便宜60%以上，支持50种不同语言，用于替代GPT-3.5版本的模型。\\n\\n4o-mini的图像分析价格和4o差不多，如果有图像分析需求还是4o更好一些。\\n\\n当前指向 gpt-4o-mini-2024-07-18","type":"text,image"},{"label":"gpt-4-turbo","value":"gpt-4-turbo","descr":"纯官方GPT4系列，支持文本和图片分析，最大回复4k，openai于2024-4-9新增的模型，知识库更新于2023年12月。提高了写作、数学、逻辑推理和编码能力。当前指向gpt-4-turbo-2024-04-09","type":"text,image"},{"label":"gpt-4-turbo-preview","value":"gpt-4-turbo-preview","descr":"纯官方GPT4系列，最大回复4k，知识库更新于2023年4月。当前指向gpt-4-0125-preview","type":"text"},{"label":"gpt-3.5-turbo-0125","value":"gpt-3.5-turbo-0125","descr":"openai于2024年1月25号更新的gpt-3.5模型，最大回复4k。\\n\\n综合能力强，过去使用最广泛的文本模型。","type":"text"},{"label":"gpt-3.5-turbo-1106","value":"gpt-3.5-turbo-1106","descr":"openai于2023年11月6号更新的gpt-3.5模型，最大回复4k。属于即将被淘汰的模型。\\n\\n建议使用gpt-3.5-turbo或gpt-4o-mini","type":"text"},{"label":"gpt-3.5-turbo-0613","value":"gpt-3.5-turbo-0613","descr":"通过微调后可以更准确地按照用户的指示进行操作，生成更简洁和针对性的输出。它不仅可以用于文本生成，还可以通过函数调用功能与其他系统和API进行集成，实现更复杂的任务自动化","type":"text"},{"label":"gpt-4o-2024-05-13","value":"gpt-4o-2024-05-13","descr":"GPT-4o，是openai的新旗舰型号，支持文本和图片分析。\\n\\n是迈向更自然的人机交互的一步——它接受文本和图像的任意组合作为输入，并生成文本和图像输出的任意组合。\\n\\n该模型为初代的4o模型","type":"text,image"},{"label":"gpt-4-turbo-2024-04-09","value":"gpt-4-turbo-2024-04-09","descr":"纯官方GPT4系列，支持文本和图片分析，最大回复4k，openai于2024-4-9新增的模型，提高了写作、数学、逻辑推理和编码能力。知识库更新于2023年12月。","type":"text,image"},{"label":"gpt-4-0125-preview","value":"gpt-4-0125-preview","descr":"纯官方GPT4系列，最大回复4k，知识库更新于2023年4月。当前与gpt-4-turbo-preview属于同一模型","type":"text"},{"label":"gpt-4-1106-preview","value":"gpt-4-1106-preview","descr":"纯官方GPT4系列，最大回复4k，知识库更新于2023年4月。正在逐渐被新的模型gpt-4-turbo和gpt-4-turbo-preview取代。","type":"text"}],"EMBED":[{"label":"text-embedding-ada-002","value":"text-embedding-ada-002","descr":"用于生成文本嵌入的模型。文本嵌入是将文本转换为数值形式（通常是向量），以便可以用于机器学习模型。","type":"vector,embeddings"},{"label":"text-embedding-3-small","value":"text-embedding-3-small","descr":"用于生成文本的嵌入表示，网络结构较小，计算资源需求较低。虽然可能不如\\"large\\"版本那样精准，但它更适合于资源受限的环境或需要更快速处理的任务。","type":"vector,embeddings"},{"label":"text-embedding-3-large","value":"text-embedding-3-large","descr":"用于生成文本的嵌入表示，即将文本转换为高维空间中的点，这些点的距离可以表示文本之间的相似度。有较大的网络结构，能够捕捉更丰富的语言特征，适用于需要高质量文本相似度或分类任务的场景。","type":"vector,embeddings"}],"IMAGE":[{"label":"gpt-image-1.5","value":"gpt-image-1.5","descr":"openAI强大的图片生成模型，支持图像生成与编辑功能，生图图片支持中文文字。相比上一版本，拥有更好的指令跟踪和对提示词的遵循。","type":"imageGen"},{"label":"gpt-image-1","value":"gpt-image-1","descr":"是一个图片生成模型，支持图像生成与编辑功能，生图图片支持中文文字。","type":"imageGen"},{"label":"dall-e-2","value":"dall-e-2","descr":"是OpenAI推出的文本生成图像系统","type":"imageGen"},{"label":"dall-e-3","value":"dall-e-3","descr":"是OpenAI开发的第三代文本到图像生成模型,提升了图像细节处理和艺术风格迁移能力","type":"imageGen"}],"type":["LLM","EMBED","IMAGE"],"baseUrl":"https://api.openai.com/v1/","LLMDefaultValue":"gpt-4o-mini","EMBEDDefaultValue":"text-embedding-ada-002","IMAGEDefaultValue":"gpt-image-1.5"},{"title":"通义千问","value":"QWEN","LLM":[{"label":"qwen-turbo","value":"qwen-turbo","descr":"通义千问超大规模语言模型，支持中文、英文等不同语言输入。适合文本创作、文本处理、编程辅助、翻译服务、对话模拟。","type":"text"},{"label":"qwen-plus","value":"qwen-plus","descr":"通义千问超大规模语言模型，支持中文、英文等不同语言输入。适合文本创作、文本处理、编程辅助、翻译服务、对话模拟。","type":"text"},{"label":"qwen-max","value":"qwen-max","descr":"暂无描述内容！","type":"text"}],"EMBED":[{"label":"text-embedding-v2","value":"text-embedding-v2","descr":"是一种将文本数据转换为向量的技术，通过深度学习模型将文本的语义信息嵌入到高维向量空间中。这些向量不仅能表达文本内容，还能捕捉文本之间的相似性和关系，从而让计算机高效地进行文本检索、分类、聚类等任务。","type":"vector"}],"IMAGE":[{"label":"wan2.2-t2i-plus","value":"wan2.2-t2i-plus","descr":"全新升级的通义万相2.2文生图，更快的生成速度。在生成图像创意性、稳定性、写实质感方面全面升级，指令遵循更强，原生支持多种风格。支持最大200万像素生成，支持智能提示词改写等。","type":"imageGen"},{"label":"wan2.2-t2i-flash","value":"wan2.2-t2i-flash","descr":"全新升级的通义万相2.2文生图，更快的生成速度。在生成图像创意性、稳定性、写实质感方面全面升级，指令遵循更强，原生支持多种风格。支持最大200万像素生成，支持智能提示词改写等。","type":"imageGen"},{"label":"wan2.5-i2i-preview","value":"wan2.5-i2i-preview","descr":"是万相2.5系列中的图像编辑模型，专注于通用图像编辑任务，支持单图编辑和多图融合功能","type":"imageGen"},{"label":"wanx2.1-imageedit","value":"wanx2.1-imageedit","descr":"是阿里云百炼平台提供的通用图像编辑模型，适用于多种图像处理任务，如风格迁移、内容编辑、尺寸优化等","type":"imageGen"},{"label":"wanx-v1","value":"wanx-v1","descr":"是通义万相系列中的文本生成图像模型，支持中英文双语输入和多种图像风格生成","type":"imageGen"}],"type":["LLM","EMBED","IMAGE"],"baseUrl":"https://dashscope.aliyuncs.com/api/v1/services/","baseImageUrl":"https://dashscope.aliyuncs.com/api/v1/","LLMDefaultValue":"qwen-plus","EMBEDDefaultValue":"text-embedding-v2","IMAGEDefaultValue":"wan2.2-t2i-plus"},{"title":"千帆大模型","value":"QIANFAN","LLM":[{"label":"ERNIE-Bot","value":"ERNIE-Bot","descr":"是百度推出的一款知识增强大语言模型，主要用于与人对话互动、回答问题、协助创作，帮助人们高效便捷地获取信息、知识和灵感","type":"text"},{"label":"ERNIE-Bot 4.0","value":"ERNIE-Bot 4.0","descr":"百度自行研发的文心产业级知识增强大语言模型4.0版本\\n\\n实现了基础模型的全面升级，在理解、生成、逻辑和记忆能力上相对ERNIE 3.5都有着显著提升，支持5K输入+2K输出。","type":"text"},{"label":"ERNIE-Bot-8K","value":"ERNIE-Bot-8K","descr":"主要用于数据分析场景，特别是在企业数据分析中表现出色。ERNIE-Bot-8K是百度文心大模型的一个版本，具有模型效果优、生成能力强、应用门槛低等独特优势。","type":"text"},{"label":"ERNIE-Bot-turbo","value":"ERNIE-Bot-turbo","descr":"是一个大语言模型，主要用于对话问答、内容创作生成等任务。它是百度自行研发的大语言模型，覆盖了海量中文数据，具有更强的对话问答和内容创作生成能力","type":"text"},{"label":"ERNIE-Speed-128K","value":"ERNIE-Speed-128K","descr":"是一款基于Transformer结构的轻量级语言模型，旨在满足实时数据处理的需求。它具有高效、低延迟和高准确性的特点，广泛应用于自然语言处理、信息检索和文本分类等领域","type":"text"},{"label":"EB-turbo-AppBuilder","value":"EB-turbo-AppBuilder","descr":"主要用于企业级应用场景，如智能客服、内容创作和知识问答等任务。它是基于文心高性能大语言模型ERNIE-Bot-turbo构建的，针对企业特定需求进行了深度的场景效果优化和输出格式定制，因此在满足企业特定需求方面具有更高的灵活性和实用性","type":"text"},{"label":"Yi-34B-Chat","value":"Yi-34B-Chat","descr":"Yi-34B-Chat是一款基于Transformer架构的生成式预训练语言模型，它拥有340亿个参数，使其在处理自然语言任务时表现出了强大的能力。","type":"text"},{"label":"BLOOMZ-7B","value":"BLOOMZ-7B","descr":"是一个用于生成文本序列的自回归模型，它可以进行多语言处理，支持46种语言和13种编程语言。BLOOMZ-7B是BLOOM模型的一个调优版本，具有更出色的泛化和零样本学习能力，适用于多种任务和场景","type":"text"},{"label":"Qianfan-BLOOMZ-7B-compressed","value":"Qianfan-BLOOMZ-7B-compressed","descr":"是千帆团队在BLOOMZ-7B基础上的压缩版本，融合量化、稀疏化等技术，显存占用降低30%以上。","type":"text"},{"label":"Mixtral-8x7B-Instruct","value":"Mixtral-8x7B-Instruct","descr":"由Mistral AI发布的首个高质量稀疏专家混合模型 (MOE)，模型由8个70亿参数专家模型组成，在多个基准测试中表现优于Llama-2-70B及GPT3.5，能够处理32K上下文，在代码生成任务中表现尤为优异。","type":"text"},{"label":"Llama-2-7b-chat","value":"Llama-2-7b-chat","descr":"由Meta AI研发并开源，在编码、推理及知识应用等场景表现优秀，Llama-2-7b-chat是高性能原生开源版本，适用于对话场景。","type":"text"},{"label":"Llama-2-13b-chat","value":"Llama-2-13b-chat","descr":"由Meta AI研发并开源，在编码、推理及知识应用等场景表现优秀，Llama-2-13b-chat是性能与效果均衡的原生开源版本，适用于对话场景。","type":"text"},{"label":"Llama-2-70b-chat","value":"Llama-2-70b-chat","descr":"由Meta AI研发并开源，在编码、推理及知识应用等场景表现优秀，Llama-2-70b-chat是高精度效果的原生开源版本。","type":"text"},{"label":"Qianfan-Chinese-Llama-2-7B","value":"Qianfan-Chinese-Llama-2-7B","descr":"是千帆团队在Llama-2-7b基础上的中文增强版本，在CMMLU、C-EVAL等中文数据集上表现优异。","type":"text"},{"label":"ChatGLM2-6B-32K","value":"ChatGLM2-6B-32K","descr":"是在ChatGLM2-6B的基础上进一步强化了对于长文本的理解能力，能够更好的处理最多32K长度的上下文。","type":"text"},{"label":"AquilaChat-7B","value":"AquilaChat-7B","descr":"是由智源研究院研发，基于Aquila-7B训练的对话模型，支持流畅的文本对话及多种语言类生成任务，通过定义可扩展的特殊指令规范，实现 AquilaChat对其它模型和工具的调用，且易于扩展。","type":"text"}],"EMBED":[{"label":"Embedding-V1","value":"Embedding-V1","descr":"主要用于将离散对象（如文本、图像等）映射为连续的数值向量，以便于计算机处理和机器学习模型的训练和使用","type":"vector,embeddings"},{"label":"tao-8k","value":"tao-8k","descr":"是由Huggingface开发者amu研发并开源的长文本向量表示模型,支持8k上下文长度,模型效果在C-MTEB上居前列,是当前最优的中文长文本embeddings模型之一","type":"vector"},{"label":"bge-large-zh","value":"bge-large-zh","descr":"是由智源研究院研发的中文版文本表示模型，可将任意文本映射为低维稠密向量，以用于检索、分类、聚类或语义匹配等任务，并可支持为大模型调用外部知识。","type":"vector"},{"label":"bge-large-en","value":"bge-large-en","descr":"是由智源研究院研发的英文版文本表示模型，可将任意文本映射为低维稠密向量，以用于检索、分类、聚类或语义匹配等任务，并可支持为大模型调用外部知识。","type":"vector"}],"type":["LLM","EMBED"],"baseUrl":"https://aip.baidubce.com","LLMDefaultValue":"Yi-34B-Chat","EMBEDDefaultValue":"Embedding-V1"},{"title":"智谱AI","value":"ZHIPU","LLM":[{"label":"glm-4","value":"glm-4","descr":"是一个多模态大语言模型，主要用于处理复杂的指令和任务，支持长文本处理、多模态理解和文生图等功能","type":"text,image"},{"label":"glm-4v","value":"glm-4v","descr":"智谱：多模态模型\\n\\n更懂中文的视觉理解、文生图等多模态模型能力。准确理解各任务场景语言描述及指令，更精确的完成多模态理解类任务，或生成高质量的图片、视频等多模态内容。","type":"text,image"},{"label":"glm-4-flash","value":"glm-4-flash","descr":"该模型官方免费，主要用于处理多种自然语言处理任务，包括智能对话助手、辅助论文翻译、ppt及会议内容生产、网页智能搜索、数据生成和抽取、网页解析、智能规划和决策、辅助科研等场景","type":"text"},{"label":"glm-3-turbo","value":"glm-3-turbo","descr":"是一种基于transformer结构的语言模型，由智谱AI推出。其主要特点包括使用三层transformer结构、采用Turbo机制以实时生成文本、处理长文本输入并具有强大的语言理解能力","type":"text"}],"EMBED":[{"label":"Embedding-3","value":"Embedding-3","descr":"主要用于文本搜索、聚类、推荐等任务。它通过将文本映射到低维向量空间，使得文本之间的语义关系可以通过向量之间的距离或相似度来衡量，从而支持各种基于向量的应用。","type":"vector"},{"label":"Embedding-2","value":"Embedding-2","descr":"用于将高维离散数据映射到低维连续数值向量中，以便机器学习模型能够更好地处理和理解这些数据","type":"vector"}],"IMAGE":[{"label":"CogView-4","value":"CogView-4","descr":"智谱首个支持生成汉字的开源文生图模型，在语义理解、图像生成质量、中英文字生成能力等方面全面提升，支持任意长度的中英双语输入，能够生成在给定范围内的任意分辨率图像。","type":"imageGen"},{"label":"Cogview-3-Flash","value":"Cogview-3-Flash","descr":"是智谱推出的免费图像生成模型，能够根据用户指令生成符合要求且美学评分更高的图像。CogView-3-Flash 主要应用于艺术创作、设计参考、游戏开发、虚拟现实等领域，帮助用户快速实现从文本到图像的转换需求。","type":"imageGen"}],"type":["LLM","EMBED","IMAGE"],"baseUrl":"https://open.bigmodel.cn","LLMDefaultValue":"glm-4-flash","EMBEDDefaultValue":"Embedding-2","IMAGEDefaultValue":"CogView-4"}]'),S={data:we},Ce="/assets/anthropic-CmeYjUz9.png",Ie="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAB4AAAAeCAYAAAA7MK6iAAAACXBIWXMAAA7DAAAOwwHHb6hkAAAAGXRFWHRTb2Z0d2FyZQB3d3cuaW5rc2NhcGUub3Jnm+48GgAAAyhJREFUSIntlk1slFUUhp/3Tv+1RmOKFTUiGBJ/kIWJC42LsiJtmQkaWBCqUSSmnakGG+MOwsINiyqtU+iKiMCidWHHaPyJdqELXNAmLhoTgsoGFmCJ1Jp2fr7XRTt0hk4LrRoWcFZfzvnOec49973f/WSbW2HhllDvgO+A/0+rWk3SziR3z4hWFD1NFH7MpPmmGNuW9Lf1a9Q2dIDscjW03DmOd/OYHHXaehHRDNQAF4AngQZgsi6n9UOD/AmwvZN1hZh/k3x8tjrs+6qXyWKtxB4aaWDrSD/Dy644nqQLudeoFpWF1pY83zUb41HgZ4CcCAGw9Upt1lu2vqPN1bO0S9Ez1CuBPQZhGJbY40SK15HTQO2S45izWgefTiR5CaDhMufnJ4Lh4Zqsf5H8MagHeFwK3xUTF4HbkjQbf3gDYKnVWx7elmTv0BAFSe/NcQFouvaWGb/6F8eXBMeIUkDjCsAAQfLReJI3R/o5Ievdktgk+EhdXi2jx5gpOheJK5HyuM2MgnrnGo02YcWBzcuAo4VF+BPEFFYX8MfnaTXZLFLwInEZ1iO+LqpvXgz7491swe4DnipLkAcUhQ8sHwZaQR0lmAuVoFBZXDXMHZcyy/TzfT7oWeGT5Z3qCYh6yGkHMHFdUz9Ugi4FvghsbEvSfH3gyz5mM+nQAT5V4m6xtGlqLVnQwXJuuHmw8U9AdRXsrZRg46npsAcYK3G/0HjJv4IHSnyXnSNz0+BAGAFwcGfrW9xTKWn0GDMhqAMWVAo8Aty/sAAdzgzy902DrzbxKXAO82BVIUpLC9+t9i7u297JOoDP+pjA6qlUVOJMfROHloJWBI8eIG/pDaCAtLs9FQ22vEYdQCzGhkLMZxOp6NDOg9Rk0gxgnyincjFvvbzqSyKRotM4DQgYw3q7Os+5XLXHgQeAscjaPb2Gs42XfArYMZ96BXmw4HD0i484v2IwQLybXdhHoPJeA1lgRPh3S89hHgI2AJHkkyP94dVVgaF41UX7QbuofGlcAffXNYX3bzTeFYGvNbCPe/M5nhdsBGplph2YIMvp5dT7r8H/td1+P3u3H/gf+dA5stOC/XQAAAAASUVORK5CYII=",Le="/assets/ollama-Bt9O-2K_.png",ke="data:image/png;base64,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",Ge="data:image/png;base64,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",Oe="/assets/qianwen-DcU66u7N.png",qe="data:image/png;base64,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",De=[{label:"id",field:"id",component:"Input",show:!1},{label:"模型名称",field:"name",required:!0,component:"Input"},{label:"模型类型",field:"modelType",slot:"modelType",required:!0,component:"Select"},{label:"基础模型",field:"modelName",required:!0,slot:"modelName",component:"Select"},{label:"API域名",field:"baseUrl",required:!0,component:"Input"},{label:"API Key",field:"apiKey",required:!0,component:"InputPassword",componentProps:{autocomplete:"new-password"},ifShow:({values:l})=>l.provider!=="OLLAMA"},{label:"Secret Key",field:"secretKey",required:!0,component:"InputPassword",ifShow:({values:l})=>!(l.provider==="DEEPSEEK"||l.provider==="OLLAMA"||l.provider==="OPENAI"||l.provider==="ZHIPU"||l.provider==="QWEN"||l.provider==="ANTHROPIC")},{label:"供应者",field:"provider",component:"Input",show:!1}],Fe=m({ANTHROPIC:Ce,DEEPSEEK:Ie,OLLAMA:Le,OPENAI:ke,QIANFAN:Ge,QWEN:Oe,ZHIPU:qe});const ya=l=>N.get({url:"/airag/airagModel/list",params:l},{isTransformResponse:!1}),Ue=l=>N.get({url:"/airag/airagModel/queryById",params:l},{isTransformResponse:!1}),Re=l=>N.post({url:"/airag/airagModel/add",params:l}),Ne=l=>N.put({url:"/airag/airagModel/edit",params:l}),Te=l=>N.post({url:"/airag/airagModel/test",params:l,timeout:120*1e3}),fa=(l,e)=>{fe.confirm({title:"确认删除",content:"是否删除名称为"+l.name+"的模型吗？",okText:"确认",cancelText:"取消",onOk:()=>N.delete({url:"/airag/airagModel/delete",params:l},{joinParamsToUrl:!0}).then(()=>{e()})})},{createMessage:Ke,createConfirm:Ea}=Ae(),Pe={name:"AddModelModal",components:{BasicForm:he,BasicModal:ge,AiModelSeniorForm:Me,AutoComplete:Ee},emits:["success","register"],setup(l,{emit:e}){const s=m([]),a=m([]),k=m(!1),P=m("all"),B=m([]),d=m("list"),v=m(""),h=m([]),q=m([]),f=m({}),D=m(1),C=m({}),G=m(!1),O=m(),T=m(!1),x=m(!1),V=t=>Fe.value[t],Q=(t,o)=>o.value.toUpperCase().indexOf(t.toUpperCase())>=0,[F,{resetFields:Z,setFieldsValue:r,validate:w,clearValidate:E}]=xe({schemas:De,showActionButtonGroup:!1,layout:"vertical",wrapperCol:{span:24}}),[_,{closeModal:K,setModalProps:A}]=be(t=>R(null,null,function*(){if(D.value=1,G.value=!1,d.value!=="list"&&(yield Z()),A({minHeight:500}),t.id){d.value="edit";let o=yield Ue({id:t.id});if(o){if(o.result.credential){let U=JSON.parse(o.result.credential);U.secretKey&&(o.result.secretKey=U.secretKey),U.apiKey&&(o.result.apiKey=U.apiKey)}let I=o.result.provider,L=S.data.filter(U=>U.value.includes(I));L&&L.length>0&&(h.value=L[0].type,q.value=L[0][o.result.modelType]),o.result.modelType&&o.result.modelType==="LLM"&&(G.value=!0),o.result.activateFlag?x.value=!0:x.value=!1,o.result.modelParams&&(C.value=JSON.parse(o.result.modelParams)),k.value=!0,yield r($({},o.result)),z()}}else k.value=!1,z(),d.value="list",q.value=[],x.value=!1}));ae();function ae(){ve("model_type").then(t=>{a.value=Be(t),t[0].value!="all"&&a.value.unshift({text:"全部类型",value:"all"})})}function te(t){if(t=="all"){B.value=S.data;return}let o=S.data.filter(I=>I.type.includes(t));B.value=o}function z(){B.value=S.data}function le(t){d.value="add",f.value=t,v.value=t.title,h.value=t.type,setTimeout(()=>{r({provider:t.value,baseUrl:t.baseUrl})},100)}function W(){return R(this,null,function*(){try{A({confirmLoading:!0});let t=yield w(),o={apiKey:t.apiKey,secretKey:t.secretKey};if(O.value){let I=O.value.emitChange();I&&(t.modelParams=JSON.stringify(I))}x.value?t.activateFlag=1:t.activateFlag=0,t.credential=JSON.stringify(o),t.id?(yield Ne(t),K(),e("success")):(t.provider=f.value.value,yield Re(t),K(),e("success"))}catch(t){t.hasOwnProperty("errorFields")&&(D.value=1)}finally{A({confirmLoading:!1})}})}function ne(){d.value="list",K(),e("success")}function oe(t=!1){return R(this,null,function*(){try{T.value=!0;let o=yield w(),I={apiKey:o.apiKey,secretKey:o.secretKey};if(O.value){let L=O.value.emitChange();L&&(o.modelParams=JSON.stringify(L))}o.credential=JSON.stringify(I),o.provider||(o.provider=f.value.value),yield Te(o).then(L=>R(null,null,function*(){if(t)return Ke.success("测试连接成功"),!0;x.value=!0,yield W()}))}catch(o){o.hasOwnProperty("errorFields")&&(D.value=1)}finally{T.value=!1}})}function ie(t){return R(this,null,function*(){yield r({modelName:""}),yield E("modelName"),yield r({modelName:f.value[t+"DefaultValue"]}),q.value=f.value[t],t==="LLM"?G.value=!0:G.value=!1,t==="IMAGE"&&f.value.baseImageUrl?r({baseUrl:f.value.baseImageUrl}):f.value.baseUrl&&r({baseUrl:f.value.baseUrl})})}function re(){d.value==="add"&&(d.value="list")}function se(t){return t?t.replaceAll(`
|
||
`,"<br>"):"暂无描述内容"}return{registerModal:_,modelTypeData:s,modelTypeOption:a,modelType:P,handleChange:te,modelTypeList:B,getImage:V,handleClick:le,dataIndex:d,providerName:v,save:W,cancel:ne,registerForm:F,handleModelTypeChange:ie,modelTypeAddOption:h,modelNameAddOption:q,goToList:re,modelTypeDisabled:k,activeKey:D,modelParams:C,modelParamsShow:G,modelActivate:x,modelParamsRef:O,filterOption:Q,getTitle:se,test:oe,testLoading:T}}},He={class:"modal"},Se={class:"header"},Ve={class:"header-title"},Qe={style:{color:"#333333"},href:"https://help.jeecg.com/aigc/guide/model/#2-%E4%BE%9B%E5%BA%94%E5%95%86%E9%80%89%E6%8B%A9",target:"_blank"},Ze={key:1,class:"add-header-title"},_e={key:2,style:{color:"#1f2329"}},Xe={key:0,class:"model-content"},Je={class:"model-header"},ze={class:"flex"},We=["src"],Ye={class:"header-text"},je={style:{display:"flex"}},$e={class:"model-content"},ea={key:0},aa={key:1},ta={key:2},la=["innerHTML"],na={style:{display:"flex","justify-content":"space-between"}},oa={style:{display:"flex"}};function ia(l,e,s,a,k,P){const B=g("Icon"),d=g("a-tooltip"),v=g("a-select-option"),h=g("a-select"),q=g("a-card"),f=g("a-col"),D=g("a-row"),C=g("a-tag"),G=g("AutoComplete"),O=g("BasicForm"),T=g("a-alert"),x=g("a-tab-pane"),V=g("AiModelSeniorForm"),Q=g("a-tabs"),F=g("a-button"),Z=g("BasicModal");return i(),u(Z,{destroyOnClose:"",onRegister:a.registerModal,canFullscreen:!1,width:"600px",wrapClassName:"ai-model-modal"},pe({default:n(()=>[c("div",He,[c("div",Se,[c("span",Ve,[a.dataIndex==="list"||a.dataIndex==="add"?(i(),M("span",{key:0,class:me(a.dataIndex==="list"?"":"add-header-title pointer"),onClick:e[0]||(e[0]=(...r)=>a.goToList&&a.goToList(...r))},[e[6]||(e[6]=y(" 选择供应商 ",-1)),a.dataIndex==="list"?(i(),u(d,{key:0,title:"供应商文档"},{default:n(()=>[c("a",Qe,[b(B,{style:{position:"relative",left:"-2px",top:"1px"},icon:"ant-design:question-circle-outlined"})])]),_:1})):p("",!0)],2)):p("",!0),a.dataIndex==="add"?(i(),M("span",Ze," > ")):p("",!0),a.dataIndex==="add"?(i(),M("span",_e,"添加 "+H(a.providerName),1)):p("",!0)]),a.dataIndex==="list"?(i(),u(h,{key:0,bordered:!1,class:"header-select",size:"small",value:a.modelType,"onUpdate:value":e[1]||(e[1]=r=>a.modelType=r),onChange:a.handleChange},{default:n(()=>[(i(!0),M(X,null,J(a.modelTypeOption,r=>(i(),u(v,{value:r.value},{default:n(()=>[y(H(r.text),1)]),_:2},1032,["value"]))),256))]),_:1},8,["value","onChange"])):p("",!0)]),a.dataIndex==="list"?(i(),M("div",Xe,[b(D,{span:24},{default:n(()=>[(i(!0),M(X,null,J(a.modelTypeList,r=>(i(),u(f,{xxl:12,xl:12,lg:12,md:12,sm:12,xs:24},{default:n(()=>[b(q,{class:"model-card",onClick:w=>a.handleClick(r)},{default:n(()=>[c("div",Je,[c("div",ze,[c("img",{src:a.getImage(r.value),class:"header-img"},null,8,We),c("div",Ye,H(r.title),1)])])]),_:2},1032,["onClick"])]),_:2},1024))),256))]),_:1})])):p("",!0),a.dataIndex==="add"||a.dataIndex==="edit"?(i(),u(Q,{key:1,activeKey:a.activeKey,"onUpdate:activeKey":e[4]||(e[4]=r=>a.activeKey=r)},{default:n(()=>[(i(),u(x,{key:1},{tab:n(()=>[c("span",je,[e[7]||(e[7]=y(" 基础信息 ",-1)),b(d,{title:"基础信息文档"},{default:n(()=>[c("a",{onClick:e[2]||(e[2]=ee(()=>{},["stop"])),style:{color:"unset"},href:"https://help.jeecg.com/aigc/guide/model/#31-%E5%A1%AB%E5%86%99%E5%9F%BA%E7%A1%80%E4%BF%A1%E6%81%AF",target:"_blank"},[b(B,{style:{position:"relative",left:"2px",top:"1px"},icon:"ant-design:question-circle-outlined"})])]),_:1})])]),default:n(()=>[c("div",$e,[b(O,{onRegister:a.registerForm},{modelType:n(({model:r,field:w})=>[b(h,{value:r[w],"onUpdate:value":E=>r[w]=E,onChange:a.handleModelTypeChange,disabled:a.modelTypeDisabled},{default:n(()=>[(i(!0),M(X,null,J(a.modelTypeAddOption,E=>(i(),u(v,{value:E},{default:n(()=>[E==="LLM"?(i(),M("span",ea,"语言模型")):E==="EMBED"?(i(),M("span",aa,"向量模型")):E==="IMAGE"?(i(),M("span",ta,"图像模型")):p("",!0)]),_:2},1032,["value"]))),256))]),_:1},8,["value","onUpdate:value","onChange","disabled"])]),modelName:n(({model:r,field:w})=>[b(G,{value:r[w],"onUpdate:value":E=>r[w]=E,options:a.modelNameAddOption,"filter-option":a.filterOption},{option:n(({value:E,label:_,descr:K,type:A})=>[b(d,{placement:"right",color:"#ffffff",overlayInnerStyle:{color:"#646a73"}},{title:n(()=>[c("div",{innerHTML:a.getTitle(K)},null,8,la)]),default:n(()=>[c("div",na,[c("span",null,H(_),1),c("div",null,[A&&A.split(",").includes("text")?(i(),u(C,{key:0,color:"#E8D7C3"},{default:n(()=>[...e[8]||(e[8]=[y("文本",-1)])]),_:1})):p("",!0),A&&A.split(",").includes("imageGen")?(i(),u(C,{key:1,color:"#FFEBD3"},{default:n(()=>[...e[9]||(e[9]=[y("图像生成",-1)])]),_:1})):p("",!0),A&&A.split(",").includes("image")?(i(),u(C,{key:2,color:"#C3D9DC"},{default:n(()=>[...e[10]||(e[10]=[y("图像分析",-1)])]),_:1})):p("",!0),A&&A.split(",").includes("vector")?(i(),u(C,{key:3,color:"#D4E0D8"},{default:n(()=>[...e[11]||(e[11]=[y("向量",-1)])]),_:1})):p("",!0),A&&A.split(",").includes("embeddings")?(i(),u(C,{key:4,color:"#FFEBD3"},{default:n(()=>[...e[12]||(e[12]=[y("文本嵌入",-1)])]),_:1})):p("",!0)])])]),_:2},1024)]),_:1},8,["value","onUpdate:value","options","filter-option"])]),_:1},8,["onRegister"]),a.modelActivate?p("",!0):(i(),u(T,{key:0,message:"模型未激活，请通过下方「保存并激活」按钮激活当前模型",type:"warning","show-icon":""}))])]),_:1})),a.modelParamsShow?(i(),u(x,{key:2},{tab:n(()=>[c("span",oa,[e[13]||(e[13]=y(" 高级配置 ",-1)),b(d,{title:"高级配置文档"},{default:n(()=>[c("a",{onClick:e[3]||(e[3]=ee(()=>{},["stop"])),style:{color:"unset"},href:"https://help.jeecg.com/aigc/guide/model/#32-%E9%85%8D%E7%BD%AE%E9%AB%98%E7%BA%A7%E5%8F%82%E6%95%B0",target:"_blank"},[b(B,{style:{position:"relative",left:"2px",top:"1px"},icon:"ant-design:question-circle-outlined"})])]),_:1})])]),default:n(()=>[b(V,{ref:"modelParamsRef",modelParams:a.modelParams},null,8,["modelParams"])]),_:1})):p("",!0)]),_:1},8,["activeKey"])):p("",!0)])]),_:2},[a.dataIndex==="add"||a.dataIndex==="edit"?{name:"footer",fn:n(()=>[b(F,{onClick:a.cancel},{default:n(()=>[...e[14]||(e[14]=[y("关闭",-1)])]),_:1},8,["onClick"]),a.modelActivate?(i(),u(F,{key:0,onClick:a.test,loading:a.testLoading,type:"default"},{default:n(()=>[...e[15]||(e[15]=[y("测试连接",-1)])]),_:1},8,["onClick","loading"])):p("",!0),b(F,{onClick:a.save,type:"primary",ghost:"true"},{default:n(()=>[...e[16]||(e[16]=[y("保存",-1)])]),_:1},8,["onClick"]),a.modelActivate?p("",!0):(i(),u(F,{key:1,onClick:e[5]||(e[5]=r=>a.test(!1)),loading:a.testLoading,type:"primary"},{default:n(()=>[...e[17]||(e[17]=[y("保存并激活",-1)])]),_:1},8,["loading"]))]),key:"0"}:{name:"footer",fn:n(()=>[]),key:"1"}]),1032,["onRegister"])}const ra=ye(Pe,[["render",ia],["__scopeId","data-v-148adc85"]]),Ba=Object.freeze(Object.defineProperty({__proto__:null,default:ra},Symbol.toStringTag,{value:"Module"}));export{ra as A,Ba as a,fa as d,Fe as i,ya as l};
|