<br />
<p align="center">
<img width="150" alt="embedbasevector" src="https://user-images.githubusercontent.com/11430621/223136025-14572cac-f2aa-455c-936b-a48cb35a0c57.png">
<h1 align="center">Embedbase</h1>
<h3 align="center">Seamless data integration for LLMs</h3>
<p align="center">
<br />
<a href="https://discord.gg/pMNeuGrDky"><img alt="Discord" src="https://img.shields.io/discord/1066022656845025310?color=black&style=for-the-badge"></a>
<a href="https://badge.fury.io/py/embedbase"><img alt="PyPI" src="https://img.shields.io/pypi/v/embedbase?color=black&style=for-the-badge"></a>
<br />
<a target="_blank" href="https://colab.research.google.com/github/different-ai/embedbase/blob/main/notebooks/Embedbase_Getting_started.ipynb">
<img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/>
</a>
<p align="center">Open-source API & SDK to integrate your data and easily hook them up to LLMs</p>
<p align="center">Used by <a href="https://github.com/louis030195/obsidian-ava">AVA</a> and serving 100k request a day</p>
<div align="center">
<a href="https://app.embedbase.xyz/signup">Try the Hosted Version</a>
·
<a href="https://github.com/different-ai/embedbase/issues/new?assignees=&labels=enhancement">Request Feature</a>
·
<a href="https://github.com/different-ai/embedbase/issues/new?assignees=&labels=bug">Report Bug</a>
</div>
<br />
</p>
</p>
Check out the [docs](https://docs.embedbase.xyz) for more info.
## What is it
Embedbase is a dead-simple API to help you use [VectorDBs](https://learn.microsoft.com/en-us/semantic-kernel/concepts-ai/vectordb) and [Embeddings Models](https://en.wikipedia.org/wiki/Sentence_embedding#:~:text=Sentence%20embedding%20is%20the%20collective,to%20vectors%20of%20real%20numbers.) without needing to host them!
You can use embedbase to customize LLM (like ChatGPT!) and automatically feed them the right information.
## Installation
`npm i embedbase-js`
```js
// this examples shows how you can use embedbase to automatically add context in a ChatGPT prompt
import { createClient } from 'embedbase-js'
const question = 'What can I do with Embedbase API?'
const embedbase = createClient(
'https://api.embedbase.xyz',
'api-key')
const context = await embedbase
.dataset('my-documentation')
.createContext(question);
console.log(context)
/* [
"Embedbase API allows to store unstructured data...",
"Embedbase API has 3 main functions a) provides a plug and play solution to store embeddings b) makes it easy to connect to get the right data into llms c)..",
"Embedabase API is self-hostable...",
] */
const prompt =
`Based on the following context:\n${context.join()}\nAnswer the user's question: ${question}`
// for await allows you to stream answers
for await (const res of embedbase.generate(prompt)) {
console.log(res)
// You, can, use, ...
}
// answer:
// You can use the Embedbase API to store unstructured data and then use the data to connect it to LLMs
```
## Table of Contents
- [Getting started](#getting-started)
- [Javascript SDK](#sdk)
- [Docs and support](#docs-and-support)
- [Integrations](#our-integrations)
- [Contributing](#contributing)
## What are people building
- [Recommendation Engines: AVA uses Embedbase to help their users find related notes](https://github.com/louis030195/obsidian-ava)
- [Chat with your data: Solpilot uses Embedbase to put smart contract integration on autopilot](https://app.solpilot.xyz/chat)
- [Talk to your docs: ChatGPT-powered search for markdown documentation](https://github.com/different-ai/chat-gpt-powered-nextra)
The fastest way to get started with Embedbase is signing up for free to [Embedbase Cloud](https://app.embedbase.xyz/).
## Supported Integrations
### Connections
- [x] Any data with the sdk or api
- [x] PDF
- [ ] Github
### Vector DBs
- [x] Supabase
- [x] Postgres
- [x] Qdrant
- [ ] Weaviate
- [ ] Redis
### Embedding Models
- [x] OpenAI Embeddings
- [x] sentence-transformers
- [ ] T5
## Docs and support
Check out our [tutorials](https://docs.embedbase.xyz) for step-by-step guides, how-to's, and best practices, our documentation is powered by GPT-4, so you can ask question directly.
Ask a question in our [Discord community](https://discord.gg/pMNeuGrDky) to get support.
## Contributing
Please read [CONTRIBUTING.md](./CONTRIBUTING.md) for details on our code of conduct, and the process for submitting pull requests to us.
## Self-hosting
> Note: this render configuration works well for experimentation.
<a href="https://render.com/deploy?repo=https://github.com/different-ai/embedbase-render">
<img src="https://render.com/images/deploy-to-render-button.svg" alt="Deploy to Render">
</a>
Raw data
{
"_id": null,
"home_page": "https://github.com/different-ai/embedbase",
"name": "embedbase",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.8,<4.0",
"maintainer_email": "",
"keywords": "embeddings,machine learning,artificial intelligence,llm",
"author": "Different AI",
"author_email": "louis@embedbase.xyz",
"download_url": "https://files.pythonhosted.org/packages/1c/af/4969e51b304385798f1efc846ec4cc6f23bcda403561c2da60583d68133c/embedbase-1.2.8.tar.gz",
"platform": null,
"description": "<br />\n\n\n<p align=\"center\">\n<img width=\"150\" alt=\"embedbasevector\" src=\"https://user-images.githubusercontent.com/11430621/223136025-14572cac-f2aa-455c-936b-a48cb35a0c57.png\">\n <h1 align=\"center\">Embedbase</h1>\n\n\n<h3 align=\"center\">Seamless data integration for LLMs</h3>\n\n <p align=\"center\">\n <br />\n <a href=\"https://discord.gg/pMNeuGrDky\"><img alt=\"Discord\" src=\"https://img.shields.io/discord/1066022656845025310?color=black&style=for-the-badge\"></a>\n <a href=\"https://badge.fury.io/py/embedbase\"><img alt=\"PyPI\" src=\"https://img.shields.io/pypi/v/embedbase?color=black&style=for-the-badge\"></a>\n <br />\n <a target=\"_blank\" href=\"https://colab.research.google.com/github/different-ai/embedbase/blob/main/notebooks/Embedbase_Getting_started.ipynb\">\n <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n </a>\n <p align=\"center\">Open-source API & SDK to integrate your data and easily hook them up to LLMs</p>\n <p align=\"center\">Used by <a href=\"https://github.com/louis030195/obsidian-ava\">AVA</a> and serving 100k request a day</p>\n <div align=\"center\">\n <a href=\"https://app.embedbase.xyz/signup\">Try the Hosted Version</a>\n \u00b7\n <a href=\"https://github.com/different-ai/embedbase/issues/new?assignees=&labels=enhancement\">Request Feature</a>\n \u00b7\n <a href=\"https://github.com/different-ai/embedbase/issues/new?assignees=&labels=bug\">Report Bug</a>\n </div>\n <br />\n </p>\n</p>\n\nCheck out the [docs](https://docs.embedbase.xyz) for more info.\n\n## What is it\n\nEmbedbase is a dead-simple API to help you use [VectorDBs](https://learn.microsoft.com/en-us/semantic-kernel/concepts-ai/vectordb) and [Embeddings Models](https://en.wikipedia.org/wiki/Sentence_embedding#:~:text=Sentence%20embedding%20is%20the%20collective,to%20vectors%20of%20real%20numbers.) without needing to host them!\nYou can use embedbase to customize LLM (like ChatGPT!) and automatically feed them the right information.\n\n## Installation\n`npm i embedbase-js`\n\n```js\n// this examples shows how you can use embedbase to automatically add context in a ChatGPT prompt\nimport { createClient } from 'embedbase-js'\n\nconst question = 'What can I do with Embedbase API?'\n\nconst embedbase = createClient(\n 'https://api.embedbase.xyz',\n 'api-key')\n\nconst context = await embedbase\n.dataset('my-documentation')\n.createContext(question);\n\nconsole.log(context)\n/* [\n \"Embedbase API allows to store unstructured data...\",\n \"Embedbase API has 3 main functions a) provides a plug and play solution to store embeddings b) makes it easy to connect to get the right data into llms c)..\",\n \"Embedabase API is self-hostable...\",\n] */\n\n\nconst prompt =\n`Based on the following context:\\n${context.join()}\\nAnswer the user's question: ${question}`\n\n// for await allows you to stream answers\nfor await (const res of embedbase.generate(prompt)) {\n console.log(res)\n // You, can, use, ...\n}\n// answer:\n// You can use the Embedbase API to store unstructured data and then use the data to connect it to LLMs\n```\n\n## Table of Contents\n\n- [Getting started](#getting-started)\n- [Javascript SDK](#sdk)\n- [Docs and support](#docs-and-support)\n- [Integrations](#our-integrations)\n- [Contributing](#contributing)\n\n\n## What are people building\n\n- [Recommendation Engines: AVA uses Embedbase to help their users find related notes](https://github.com/louis030195/obsidian-ava)\n- [Chat with your data: Solpilot uses Embedbase to put smart contract integration on autopilot](https://app.solpilot.xyz/chat)\n- [Talk to your docs: ChatGPT-powered search for markdown documentation](https://github.com/different-ai/chat-gpt-powered-nextra)\n\n\n\nThe fastest way to get started with Embedbase is signing up for free to [Embedbase Cloud](https://app.embedbase.xyz/).\n\n\n\n\n\n## Supported Integrations\n\n### Connections\n- [x] Any data with the sdk or api\n- [x] PDF\n- [ ] Github\n\n### Vector DBs\n- [x] Supabase\n- [x] Postgres\n- [x] Qdrant\n- [ ] Weaviate\n- [ ] Redis\n\n### Embedding Models\n- [x] OpenAI Embeddings\n- [x] sentence-transformers\n- [ ] T5\n\n\n\n\n## Docs and support\n\nCheck out our [tutorials](https://docs.embedbase.xyz) for step-by-step guides, how-to's, and best practices, our documentation is powered by GPT-4, so you can ask question directly. \n\nAsk a question in our [Discord community](https://discord.gg/pMNeuGrDky) to get support.\n\n## Contributing\n\nPlease read [CONTRIBUTING.md](./CONTRIBUTING.md) for details on our code of conduct, and the process for submitting pull requests to us.\n\n## Self-hosting\n\n> Note: this render configuration works well for experimentation. \n\n<a href=\"https://render.com/deploy?repo=https://github.com/different-ai/embedbase-render\">\n <img src=\"https://render.com/images/deploy-to-render-button.svg\" alt=\"Deploy to Render\">\n</a>\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Open-source API & SDK to integrate your data and easily hook them up to LLMs.",
"version": "1.2.8",
"project_urls": {
"Homepage": "https://github.com/different-ai/embedbase",
"Repository": "https://github.com/different-ai/embedbase"
},
"split_keywords": [
"embeddings",
"machine learning",
"artificial intelligence",
"llm"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "b862f7336abcaf52aa6df1144b5334e00057304905a4ec0f12ed29da90ea9470",
"md5": "75c3dfc3bee0b0955ad7bb097e817a62",
"sha256": "4bc1310795ab06fb575e48eec2092edc78313746a0f138768010511cd97f301f"
},
"downloads": -1,
"filename": "embedbase-1.2.8-py3-none-any.whl",
"has_sig": false,
"md5_digest": "75c3dfc3bee0b0955ad7bb097e817a62",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8,<4.0",
"size": 27654,
"upload_time": "2023-06-21T19:26:52",
"upload_time_iso_8601": "2023-06-21T19:26:52.968109Z",
"url": "https://files.pythonhosted.org/packages/b8/62/f7336abcaf52aa6df1144b5334e00057304905a4ec0f12ed29da90ea9470/embedbase-1.2.8-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "1caf4969e51b304385798f1efc846ec4cc6f23bcda403561c2da60583d68133c",
"md5": "f61fb3619b1f535f13cd79e93308257d",
"sha256": "4bbccbd298d09e0d16c2ec2b1309be690b79053f085d28460e807c6e96f14f03"
},
"downloads": -1,
"filename": "embedbase-1.2.8.tar.gz",
"has_sig": false,
"md5_digest": "f61fb3619b1f535f13cd79e93308257d",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8,<4.0",
"size": 23355,
"upload_time": "2023-06-21T19:26:55",
"upload_time_iso_8601": "2023-06-21T19:26:55.006890Z",
"url": "https://files.pythonhosted.org/packages/1c/af/4969e51b304385798f1efc846ec4cc6f23bcda403561c2da60583d68133c/embedbase-1.2.8.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-06-21 19:26:55",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "different-ai",
"github_project": "embedbase",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "embedbase"
}
JSON
