# How to automatically import ChatGPT conversations into Fibery

**URL:** <https://community.fibery.io/t/how-to-automatically-import-chatgpt-conversations-into-fibery/4714>\
**Category:** Get Help\
**Created:** [July 3, 2023, 1:48pm UTC](https://community.fibery.io/t/how-to-automatically-import-chatgpt-conversations-into-fibery/4714 "2023-07-03T13:48:56Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![Yuri\_BC](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/yuri_bc/32/8803_2.png) [@Yuri\_BC](https://community.fibery.io/u/Yuri_BC)\
**Post date:** [July 3, 2023, 1:48pm UTC](https://community.fibery.io/t/how-to-automatically-import-chatgpt-conversations-into-fibery/4714/1 "2023-07-03T13:48:56Z")

</div>

How to automatically import ChatGPT (GPT-4) conversations into Fibery?  
There is a chrome app that allows importing a complete conversation into Notion.so.  
Anyone found a way?

---

<div class="post-metadata">

**Author:** ![YvetteLans](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/yvettelans/32/6101_2.png) [@YvetteLans](https://community.fibery.io/u/YvetteLans)\
**Post date:** [July 3, 2023, 2:53pm UTC](https://community.fibery.io/t/how-to-automatically-import-chatgpt-conversations-into-fibery/4714/2 "2023-07-03T14:53:26Z")

</div>

Why not use Fibery’s AI chat (GPT-4) and click on insert? 🙂

Else you can use Zapier / Make / etc. or API’s.

It depends on you use case if that’s helpful. Can imagine that you only what to import a conversation if it really contains the answer you were looking for.

In my case Chappie always needs a bit time before he understands me completely.

---

<div class="post-metadata">

**Author:** ![Yuri\_BC](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/yuri_bc/32/8803_2.png) [@Yuri\_BC](https://community.fibery.io/u/Yuri_BC)\
**Post date:** [July 3, 2023, 4:18pm UTC](https://community.fibery.io/t/how-to-automatically-import-chatgpt-conversations-into-fibery/4714/3 "2023-07-03T16:18:39Z")

</div>

I’m a heavy ChatGPT user thus their monthly cost is cheaper than the OpenAI API that Fibery uses.  
Zapier, Make, etc do not have OpenAI GPT4 integrated yet as far as I can see.  
The idea is that I want a compolete chats that change to be automatically imported on click or that this just goes automatically for all chats.

---

<div class="post-metadata">

**Author:** ![YvetteLans](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/yvettelans/32/6101_2.png) [@YvetteLans](https://community.fibery.io/u/YvetteLans)\
**Post date:** [July 3, 2023, 5:27pm UTC](https://community.fibery.io/t/how-to-automatically-import-chatgpt-conversations-into-fibery/4714/4 "2023-07-03T17:27:43Z")

</div>

> [@Yuri\_BC](#):
>
> do not have OpenAI GPT4 integrated yet

[API is not available yet](https://openai.com/waitlist/gpt-4-api) so I don’t think it’s possible to import your conversations into Fibery 🤷🏻‍♀️

---

<div class="post-metadata">

**Author:** ![mdubakov](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/mdubakov/32/10_2.png) [@mdubakov](https://community.fibery.io/u/mdubakov)\
**Post date:** [July 4, 2023, 6:26am UTC](https://community.fibery.io/t/how-to-automatically-import-chatgpt-conversations-into-fibery/4714/5 "2023-07-04T06:26:31Z")

</div>

To make it work we should create ChatGPT Plugin.

---

<div class="post-metadata">

**Author:** ![Yuri\_BC](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/yuri_bc/32/8803_2.png) [@Yuri\_BC](https://community.fibery.io/u/Yuri_BC)\
**Post date:** [July 26, 2023, 12:48pm UTC](https://community.fibery.io/t/how-to-automatically-import-chatgpt-conversations-into-fibery/4714/6 "2023-07-26T12:48:33Z")

</div>

Some feedback from chatgpt 4:

**HuggingFace Integration:**

- **Versatility** : HuggingFace offers a wide range of pre-trained models, including GPT-3, GPT-4, and others, providing flexibility to choose the model that best fits your needs. [Source](https://gptaiworld.com/huggingface-vs-openai/)
- **Cost** : HuggingFace’s pricing model is based on input words, which might be more cost-effective for high request volumes compared to OpenAI’s token-based pricing. [Source](https://www.blackhatworld.com/seo/how-much-would-huggingface-cost-as-compared-to-openai-for-gpt-how-would-you-compare-the-two-pricing-models.1415853/)
- **Control and Customization** : HuggingFace allows for custom model training and deployment, providing more control over model usage. [Source](https://sourceforge.net/software/compare/Hugging-Face-vs-OpenAI/)
- **Partnerships** : HuggingFace has partnerships with major cloud providers like AWS, enabling easier integration and deployment. [Source](https://www.forbes.com/sites/janakirammsv/2023/02/22/aws-and-hugging-face-partner-to-offer-an-alternative-to-microsofts-azure-openai-service/)

**OpenAI/ChatGPT API:**

- **Specialization** : OpenAI’s ChatGPT API is designed specifically for conversational AI, providing robust and reliable conversational capabilities. [Source](https://www.forbes.com/sites/janakirammsv/2023/02/22/aws-and-hugging-face-partner-to-offer-an-alternative-to-microsofts-azure-openai-service/)
- **Ease of Integration** : OpenAI provides a managed service with a REST API, making integration simpler and more straightforward. [Source](https://www.forbes.com/sites/janakirammsv/2023/02/22/aws-and-hugging-face-partner-to-offer-an-alternative-to-microsofts-azure-openai-service/)

**Meta LLaMA:**

- **Open Source** : LLaMA 2 is open-source and free for research and commercial use, providing a cost-effective option. [Source](https://about.fb.com/news/2023/07/llama-2/)
- **Fine-Tuning and Integration** : LLaMA 2 is ready for fine-tuning on platforms like AWS, Azure, and Hugging Face’s AI model hosting platform. [Source](https://dataconomy.com/2023/07/19/meta-ai-what-is-llama-2-and-how-to-use/)
- **Architectural Improvements** : LLaMA 2 includes architectural tweaks like Grouped Query Attention and is pre-trained on a large corpus of 2 trillion tokens. [Source](https://huggingface.co/docs/transformers/main/en/model_doc/llama)

In conclusion, the choice between HuggingFace, OpenAI, and Meta LLaMA depends on your specific needs, such as the level of control, cost-effectiveness, ease of integration, and the specific AI capabilities you require.

---

<div class="post-metadata">

**Author:** ![Illusory](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/illusory/32/10403_2.png) [@Illusory](https://community.fibery.io/u/Illusory)\
**Post date:** [July 28, 2023, 1:36pm UTC](https://community.fibery.io/t/how-to-automatically-import-chatgpt-conversations-into-fibery/4714/7 "2023-07-28T13:36:34Z")

</div>

We would also greatly appreciate the ability to sync ChatGPT messages to Fibery.

Another potential addon would be related to furthering the development of the Fibery Extension or a standalone Fibery \<\> ChatGPT extension to support saving ChatGPT messages.

We don’t use Notion (\*anymore), though the thought had crossed my mind to use this extension and Sync Notion \> Fibery purely for this.

> **[ChatGPT to Notion](https://chrome.google.com/webstore/detail/chatgpt-to-notion/oojndninaelbpllebamcojkdecjjhcle)**
>
> Save your discussions with ChatGPT directly into Notion!

Not as great as native sync, but may be helpful or useful for many, and I imagine it may be faster to support ChatGPT history saving to Fibery quicker than a native plugin.
