LLMSERP
searchALL
imageIMAGES
videocamVIDEOS
shopping_cartSHOPPING
TOOLS
About 12 results (0.41 seconds)
Jina AI is an open-source neural search company. We build the next generation search for unstructured data. Our mission is to enable developers to build AI-powered search applications.
I'm sorry, but there is no Wikipedia article about "Jina AI". Perhaps you meant the name "Jina" which is a name of Indian origin.
Jina AI - Build Neural Search Applications with Python. In this video, we will build a neural search application with Jina AI.
Jina AI, a San Francisco-based startup, has raised $30 million in a series B funding round. The round was led by GGV Capital and saw participation from Mango Capital, SAP.iO Fund, and others. Jina AI offers an open-source neural search platform that lets developers build search applications for unstructured data.
Jina AI is a San Francisco-based company that offers a neural search platform for developers. The company's platform allows developers to build search applications for unstructured data.
Read writing from Jina AI on Medium. Every day, Jina AI and thousands of other voices read, write, and share important stories on Medium.
Learn about working at Jina AI. Join LinkedIn today for free. See who you know at Jina AI, leverage your professional network, and get hired.
Jina is a neural search framework that empowers developers to build search applications for unstructured data.
Jina AI (@JinaAI_). We build the next generation search for unstructured data. Open source neural search. San Francisco, CA. jina.ai Joined July 2019.
LLMSERP
1
2
3
4
5
6
7
8
9
10
navigate_next
question_answer
什么是 SERP？
keyboard_arrow_down
question_answer
那么所有搜索结果都是由 LLM 生成的吗？
keyboard_arrow_down
question_answer
为什么搜索结果中有些 URL 损坏并导致 404 页面？
keyboard_arrow_down
question_answer
那么为什么有些 URL 是真实的并且会引导至实际的页面？
keyboard_arrow_down
question_answer
我可以通过 API 调用它吗？
keyboard_arrow_down
question_answer
但是如果所有搜索结果都是假的，那么这个演示或 API 有什么意义呢？
keyboard_arrow_down
question_answer
它是开源的吗？
keyboard_arrow_down
速率限制
速率限制通过三种方式跟踪：RPM（每分钟请求数）和TPM（每分钟词元数）。限制按 IP/API 密钥强制执行，当首先达到 RPM 或 TPM 阈值时，将触发限制。当您在请求标头中提供 API 密钥时，我们会按密钥而不是 IP 地址跟踪速率限制。
| 产品 | API端口 | 描述arrow_upward | 无 API 密钥key_off | 使用 API 密钥key | 带有高级 API 密钥key | 平均延迟 | 词元使用计数 | 请求类型 | |
|---|---|---|---|---|---|---|---|---|---|
| 读取器 API | https://r.jina.ai | 将 URL 转换为大模型友好文本 | 20 RPM | 500 RPM | trending_up5000 RPM | 7.9s | 以输出响应中的词元数量为准。 | GET/POST | |
| 读取器 API | https://s.jina.ai | 搜索网络并将结果转换为大模型友好文本 | block | 100 RPM | trending_up1000 RPM | 2.5s | 每个请求都需要固定数量的词元，从 10000 个词元开始 | GET/POST | |
| 深度搜索 | https://deepsearch.jina.ai/v1/chat/completions | 推理、搜索和迭代以找到最佳答案 | block | 50 RPM | 500 RPM | 56.7s | 统计整个过程中词元的总数。 | POST | |
| 向量模型API | https://api.jina.ai/v1/embeddings | 将文本/图片转为定长向量 | block | 500 RPM & 1,000,000 TPM | trending_up2,000 RPM & 5,000,000 TPM | ssid_chart 取决于输入大小 help | 以输入请求中的词元数量为准。 | POST | |
| 重排器 API | https://api.jina.ai/v1/rerank | 按查询对文档进行精排 | block | 500 RPM & 1,000,000 TPM | trending_up2,000 RPM & 5,000,000 TPM | ssid_chart 取决于输入大小 help | 以输入请求中的词元数量为准。 | POST | |
| 分类器 API | https://api.jina.ai/v1/train | 使用训练样本训练分类器 | block | 20 RPM & 200,000 TPM | 60 RPM & 1,000,000 TPM | ssid_chart 取决于输入大小 | 词元计数为：输入词元 × 迭代次数 | POST | |
| 分类器 API (少量样本) | https://api.jina.ai/v1/classify | 使用经过训练的少样本分类器对输入进行分类 | block | 20 RPM & 200,000 TPM | 60 RPM & 1,000,000 TPM | ssid_chart 取决于输入大小 | 词元计数为：输入词元 | POST | |
| 分类器 API (零样本) | https://api.jina.ai/v1/classify | 使用零样本分类对输入进行分类 | block | 200 RPM & 500,000 TPM | 1,000 RPM & 3,000,000 TPM | ssid_chart 取决于输入大小 | 词元计数为：输入词元 加 标签词元 | POST | |
| 切分器 API | https://api.jina.ai/v1/segment | 对长文本进行分词分句 | 20 RPM | 200 RPM | 1,000 RPM | 0.3s | 词元不计算使用量。 | GET/POST |
