From da1775ca6fea8bfd622b06ef2fba840ad5d6bb5a Mon Sep 17 00:00:00 2001 From: Vissidarte-Herman Date: Wed, 20 Dec 2023 01:56:21 +0800 Subject: [PATCH 1/4] Minor editorial updates --- README.md | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/README.md b/README.md index 96a42fd2a8..0240cbfd40 100644 --- a/README.md +++ b/README.md @@ -5,10 +5,10 @@

The AI-native database built for the next-gen Retrieval-Augmented Generation - Incredibly fast vector search +

Incredibly fast vector search

---- +

GitHub Discussions | YouTube | @@ -19,13 +19,13 @@

-Infinity is an open-source AI-native database designed to enhance retrieval-augmented generation (RAG) applications. As a natural partner to mainstream LLMs, Infinity solves primary challenges faced by B2B applications, such as internal enterprise search, industry-specific search, in-house AI assistants, chatbots, in-house knowledge management systems, and more. +Infinity is an open-source AI-native database designed to enhance retrieval-augmented generation (RAG) applications. As a natural partner to mainstream LLMs, Infinity solves primary challenges faced by B2B applications, such as internal enterprise search, industry-specific search, in-house AI assistants, chatbots, in-house knowledge management systems, and more. Infinity empowers these applications by supporting full-text search, multi-embedding search, multiple-collection query, and fused search. -Infinity empowers these applications by supporting full-text search, multi-embedding search, multiple-collection query, and fused search. +The vector search performance of Infinity on common datasets is exceptionally superior to all known open source vector databases, the higher the dimensionality the embeddings, the greater performance improvements achieved. -Infinity was released under the [open-source Apache License 2.0](https://github.com/infiniflow/infinity/blob/master/LICENSE) on December 21, 2023. +Infinity was released under the [open-source Apache License 2.0](https://github.com/infiniflow/infinity/blob/master/LICENSE) on December 20, 2023. -###🧐 Benchmark +### 🧐 Benchmark See a Benchmark report [here](). @@ -48,7 +48,7 @@ See [Build from Source](build_from_source.md). Infinity comes with **performance**, **flexibility**, **ease-of-use**, and many features designed to address the challenges facing the next-gen RAG applications: -### Blazing fast +### Incredibly fast - End-to-end latency as low as 0.1 ms. See the [Benchmarking](https://www.example.com). - 10K QPS on CPU: @@ -59,7 +59,7 @@ Infinity comes with **performance**, **flexibility**, **ease-of-use**, and many In addition to hybrid search, Infinity takes over the decision-making process previously owned by the upper-level applications, thereby simplifying complex queries considerably. - Full-text search -- KNN-based vector search +- KNN-based vector search ### Rich data types From db4065b54d2d027d1dce3ce49546b86cec107fcb Mon Sep 17 00:00:00 2001 From: Vissidarte-Herman Date: Wed, 20 Dec 2023 14:26:54 +0800 Subject: [PATCH 2/4] Updated readme --- README.md | 67 ++++++++++++++++++------------------------------------- 1 file changed, 22 insertions(+), 45 deletions(-) diff --git a/README.md b/README.md index 0240cbfd40..9df86e4e0c 100644 --- a/README.md +++ b/README.md @@ -4,82 +4,64 @@

- The AI-native database built for the next-gen Retrieval-Augmented Generation -

Incredibly fast vector search

- + The AI-native database built for LLM applications, offering incredibly fast vector and full-text search

GitHub Discussions | - YouTube | Roadmap 2024 | Twitter | - Apache 2.0 | Discord | + YouTube |

-Infinity is an open-source AI-native database designed to enhance retrieval-augmented generation (RAG) applications. As a natural partner to mainstream LLMs, Infinity solves primary challenges faced by B2B applications, such as internal enterprise search, industry-specific search, in-house AI assistants, chatbots, in-house knowledge management systems, and more. Infinity empowers these applications by supporting full-text search, multi-embedding search, multiple-collection query, and fused search. +Infinity is a cutting-edge AI-native database that provides a wide range of search capabilities for rich data types such as vectors, full-text, and structured data. It provides robust support for various LLM applications, including search, recommendations, question-answering, conversational AI, copilot, content generation, data management, and much RAG (Retrieval-augmented Generation) applications The vector search performance of Infinity on common datasets is exceptionally superior to all known open source vector databases, the higher the dimensionality the embeddings, the greater performance improvements achieved. Infinity was released under the [open-source Apache License 2.0](https://github.com/infiniflow/infinity/blob/master/LICENSE) on December 20, 2023. -### 🧐 Benchmark - -See a Benchmark report [here](). - -### Clients - -- [Python client]() - - -## 🎮 Get Started +## 🌟 Key Features -CONTENT MISSING HERE +Infinity comes with **performance**, **flexibility**, **ease-of-use**, and many features designed to address the challenges facing the next-gen RAG applications: -## 🛠️ Build from Source +### Incredibly fast -See [Build from Source](build_from_source.md). +0.1 milliseconds query latency with 10K QPS on million-scale vector datasets. See the [Benchmarking](https://www.example.com). +### Fused search -## 🌟 Key Features +Supports a fused search of multi-embeddings and full text, in addition to filtering. -Infinity comes with **performance**, **flexibility**, **ease-of-use**, and many features designed to address the challenges facing the next-gen RAG applications: -### Incredibly fast +### Rich data types -- End-to-end latency as low as 0.1 ms. See the [Benchmarking](https://www.example.com). -- 10K QPS on CPU: +Supports a wide range of data types including strings, numerics, vectors, and more. +### Ease-of-use -### Fused search +- Intuitive Python API. +- A single-binary architecture with no dependencies, making deployment a breeze. -In addition to hybrid search, Infinity takes over the decision-making process previously owned by the upper-level applications, thereby simplifying complex queries considerably. +### 🧐 Benchmark -- Full-text search -- KNN-based vector search +See a Benchmark report [here](). +### Clients -### Rich data types +- [Python client]() -In addition to embeddings generated by LLMs, Infinity also stores structured and semi-structured data, offering support for mixed data type queries. -- Numeric -- String -- Float -- Date -- Time -- Geography +## 🎮 Get Started -### Easy-to-use +CONTENT MISSING HERE -- One binary to deploy -- Intuitive API: - - We carefully weighed the pros and cons of similar APIs in the market and designed our own. +## 🛠️ Build from Source +See [Build from Source](build_from_source.md). ## 📑 Roadmap @@ -91,8 +73,3 @@ In addition to embeddings generated by LLMs, Infinity also stores structured and - [Twitter](https://twitter.com/infiniflowai) - [GitHub Discussions](https://github.com/infiniflow/infinity/discussions) - [YouTube](https://www.youtube.com/@InfiniFlow-AI) - - -## 👩‍💻 Contributing - -To find out how to make a contribution to Infinity, see the [contribution guidelines](CONTRIBUTING.md). From a94226b2fb21860e349ad7cb3b024ad57b098deb Mon Sep 17 00:00:00 2001 From: Vissidarte-Herman Date: Wed, 20 Dec 2023 15:19:15 +0800 Subject: [PATCH 3/4] Revamped README --- README.md | 28 +++++++--------------------- 1 file changed, 7 insertions(+), 21 deletions(-) diff --git a/README.md b/README.md index 9df86e4e0c..d23620c37c 100644 --- a/README.md +++ b/README.md @@ -8,7 +8,6 @@

- GitHub Discussions | Roadmap 2024 | Twitter | Discord | @@ -16,43 +15,30 @@

-Infinity is a cutting-edge AI-native database that provides a wide range of search capabilities for rich data types such as vectors, full-text, and structured data. It provides robust support for various LLM applications, including search, recommendations, question-answering, conversational AI, copilot, content generation, data management, and much RAG (Retrieval-augmented Generation) applications - -The vector search performance of Infinity on common datasets is exceptionally superior to all known open source vector databases, the higher the dimensionality the embeddings, the greater performance improvements achieved. - -Infinity was released under the [open-source Apache License 2.0](https://github.com/infiniflow/infinity/blob/master/LICENSE) on December 20, 2023. +Infinity is a cutting-edge AI-native database that provides a wide range of search capabilities for rich data types such as vectors, full-text, and structured data. It provides robust support for various LLM applications, including search, recommendations, question-answering, conversational AI, copilot, content generation, and many more RAG (Retrieval-augmented Generation) applications. ## 🌟 Key Features Infinity comes with **performance**, **flexibility**, **ease-of-use**, and many features designed to address the challenges facing the next-gen RAG applications: -### Incredibly fast +### ⚡️ Incredibly fast 0.1 milliseconds query latency with 10K QPS on million-scale vector datasets. See the [Benchmarking](https://www.example.com). -### Fused search +### 🔮 Fused search Supports a fused search of multi-embeddings and full text, in addition to filtering. - -### Rich data types +### 🍔 Rich data types Supports a wide range of data types including strings, numerics, vectors, and more. -### Ease-of-use +### 🎁 Ease-of-use -- Intuitive Python API. +- Intuitive Python API. See the [Python client]() - A single-binary architecture with no dependencies, making deployment a breeze. -### 🧐 Benchmark - -See a Benchmark report [here](). - -### Clients - -- [Python client]() - ## 🎮 Get Started @@ -63,7 +49,7 @@ CONTENT MISSING HERE See [Build from Source](build_from_source.md). -## 📑 Roadmap +## 📜 Roadmap - [Infinity Roadmap 2024]() From 6afafd137dc2fd02062825455c1b8fbcebe8de8e Mon Sep 17 00:00:00 2001 From: Vissidarte-Herman Date: Wed, 20 Dec 2023 15:28:13 +0800 Subject: [PATCH 4/4] minor --- README.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index d23620c37c..19e5c20a68 100644 --- a/README.md +++ b/README.md @@ -23,7 +23,9 @@ Infinity comes with **performance**, **flexibility**, **ease-of-use**, and many ### ⚡️ Incredibly fast -0.1 milliseconds query latency with 10K QPS on million-scale vector datasets. See the [Benchmarking](https://www.example.com). +- Achieves 0.1 milliseconds query latency on million-scale vector datasets. +- Up to 10K QPS on million-scale vector datasets. +- See the [Benchmarking](https://www.example.com). ### 🔮 Fused search