File size: 2,046 Bytes
831bf20
 
 
 
 
 
 
 
 
cb6dd56
 
 
 
e270529
cb6dd56
 
 
 
6503caa
 
 
 
 
cb6dd56
1aeb577
cb6dd56
1aeb577
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
---
title: README
emoji: 💻
colorFrom: purple
colorTo: purple
sdk: static
pinned: false
---

Arc53 is a UK based company that works in AI and deploys open source projects like DocsGPT.

If you have questions or want to work with us reach out to us here [Contact us](mailto:[email protected]) or read more about [Arc53](https://arc53.com/)

## [DocsGPT](https://www.docsgpt.cloud/)

DocsGPT is our open source project that focuses on GPT-powered chat for documentation, chat with your documents.
It has few moving parts around it:
1. [DocsGPT on Github](https://github.com/arc53/DocsGPT) - Here we have frontend, backend and wiki
2. [DocsGPT-7B-mistral](https://huggingface.co/Arc53/docsgpt-7b-mistral) - Our best and newest model. Purpose optimised for RAG
3. [DocsGPT-7B](https://huggingface.co/Arc53/DocsGPT-7B) - Is our finetuned version of llama-2, is a purpose optimised model for getting answers about documentation
4. [DocsGPT-14B](https://huggingface.co/Arc53/docsgpt-14b) - Is our finetuned version of llama-2, is a purpose optimised model for getting answers about documentation
5. [DocsGPT-7B-falcon](https://huggingface.co/Arc53/docsgpt-7b-falcon) - Is our finetuned version of falcon, is a purpose optimised model for getting answers about documentation
6. [DocsGPT-40B-falcon](https://huggingface.co/Arc53/docsgpt-40b-falcon) -  Is our finetuned version of falcon, is a purpose optimised model for getting answers about documentation

Our models

| Name                                                                  | Base Model  | Requirements (or similar) |
| --------------------------------------------------------------------- | ----------- | ------------------------- |
| [Docsgpt-7b-mistral](https://huggingface.co/Arc53/docsgpt-7b-mistral)   | Mistral-7b   | 1xA10G gpu                |
| [Docsgpt-14b](https://huggingface.co/Arc53/docsgpt-14b)               | llama-2-14b | 2xA10 gpu's               |
| [Docsgpt-40b-falcon](https://huggingface.co/Arc53/docsgpt-40b-falcon) | falcon-40b  | 8xA10G gpu's              |