The OpenAI API is better than ChatGPT

Stable Discussion
23 Nov 202308:36

TLDRThe video transcript discusses the advantages of using the OpenAI API over the standard ChatGPT for tasks that require customization and programmability. It highlights the limitations of ChatGPT, such as lack of control over system prompts and chat history, and demonstrates how the API allows for more consistent and trainable interactions. The speaker uses a tool like Insomnia to make API requests, showcasing the increased control over the model version and the conversation context. The summary encourages using the API for better control and results with AI models.


  • 😀 The OpenAI API provides more flexibility and control over ChatGPT's capabilities for users with programming experience.
  • 🔧 Users can create customized versions of ChatGPT, such as a personal trainer with a unique personality, using the API.
  • 🤖 Programmability is a key advantage of the API, allowing for more consistent and trainable interactions compared to the standard ChatGPT interface.
  • 📚 The API allows for the selection of specific model versions, offering stability for ongoing projects against potential changes in newer versions.
  • 🗨️ With the API, users have control over the chat context and system prompts, which are not adjustable in the standard ChatGPT experience.
  • 🔄 The API enables developers to manage the conversation history and decide what information is relevant for the AI to remember or ignore.
  • 🛠️ Tools like Insomnia or Postman can be used to make API requests, providing a user-friendly interface for interacting with the API.
  • 📝 When using the API, developers can specify the exact messages and system prompts to guide the AI's responses more precisely.
  • 🔑 The use of an OpenAI token is required for authentication when making API requests, which can be set as a bearer token in the request header.
  • 📈 The API offers a more controlled and predictable experience, which is beneficial for building real workflows and solving specific problems.
  • 📚 Full documentation is available for the API, empowering users to understand and utilize its capabilities to the fullest.

Q & A

  • Why might someone prefer using the ChatGPT API over the regular ChatGPT service for daily tasks?

    -The ChatGPT API provides more programmability and control over the conversation context, allowing users to customize their experience and integrate the AI into their applications more effectively.

  • What is an example of a custom application created using the ChatGPT API mentioned in the script?

    -An example given is a personal trainer application that uses spicy language to motivate the user, and can adjust its tone based on user feedback.

  • How does the ChatGPT API differ from the regular ChatGPT service in terms of conversation context?

    -The API allows for a subset of the conversation history to be sent, along with a system prompt, giving users more control over the context and prompting of the AI.

  • What is the significance of being able to specify a system prompt when using the ChatGPT API?

    -Specifying a system prompt allows users to guide the AI's behavior and responses in a particular direction, which is not possible with the regular ChatGPT service.

  • Why might the full chat history sent to the OpenAI API be a limitation when using the regular ChatGPT service?

    -The full chat history may include irrelevant or outdated information that could affect the AI's response, and users have no control over what part of the history is sent.

  • What tool is mentioned in the script for making API requests, and what is its function?

    -Insomnia is mentioned as a tool for making API requests. It functions similarly to Postman, allowing users to create, send, and manage HTTP requests.

  • How does using the ChatGPT API provide more consistency and trainability compared to the regular service?

    -The API allows for more precise control over the model version and the conversation context, leading to more consistent responses and the ability to train the AI for specific tasks.

  • What is the role of the 'bearer token' when making API requests to the OpenAI API?

    -The bearer token, which can be set in the authorization header of the API request, authenticates the user and grants access to the OpenAI API services.

  • Why is it beneficial to have control over the model version when using the ChatGPT API?

    -Control over the model version ensures that the AI's behavior remains consistent for the user, regardless of updates or changes made to newer versions of the model.

  • What is the main advantage of using the ChatGPT API for building real workflows and solving real problems?

    -The main advantage is the increased control and customization over the AI's responses, allowing for better integration into specific applications and tasks.

  • How can users get started with using the ChatGPT API according to the script?

    -Users can get started by obtaining an API token from ChatGPT, using tools like Insomnia to make requests, and referring to the OpenAI documentation for detailed instructions.



🤖 Advantages of Using the ChatGPT API

The first paragraph introduces the benefits of using the ChatGPT API over the standard ChatGPT service for daily tasks. It highlights the programmability and customization available through the API, which standard ChatGPT lacks. The speaker uses the example of a friend named Dan who created a personal trainer chatbot using the API, demonstrating the ability to adjust the bot's behavior dynamically. The paragraph emphasizes the control over the chat context and system prompts, which are not accessible in the standard ChatGPT interface, and suggests that the API is preferable for building workflows and solving real problems due to its consistency and trainability.


🔧 Customizing AI Interactions with the API

The second paragraph delves into the technical aspects of making API requests using a tool like Insomnia, which is likened to Postman. It explains the process of setting up an HTTP POST request to the completions endpoint and the advantages of having control over the model, model version, and system prompts. The speaker illustrates how the API allows for precise control over the chat context and the messages sent to the AI, enabling users to tailor the AI's responses more effectively. The paragraph concludes by encouraging users to explore the API's full documentation and to utilize the API for more controlled and refined interactions with AI models.




The OpenAI API refers to the application programming interface provided by OpenAI, which allows developers to integrate AI capabilities into their applications. In the video, the API is highlighted as a superior choice for tasks requiring customization and programmability, compared to using ChatGPT directly.


ChatGPT is a language model developed by OpenAI that can generate human-like text based on the prompts it receives. The video discusses the limitations of using ChatGPT without the API and how the API can overcome these limitations for more advanced use cases.

💡API Call

An API call is a request made to an API to perform a specific function or retrieve data. The video script explains that with the OpenAI API, users can make API calls to utilize ChatGPT's capabilities in a more controlled and customizable manner.


Programmability refers to the ability to control or modify the behavior of a system through programming. The video emphasizes the programmability offered by the OpenAI API, which allows for greater control over the AI's responses compared to the standard ChatGPT interface.


Consistency in the context of the video refers to the reliability and predictability of the AI's responses. The API is said to provide more consistent results for users building workflows and solving problems with AI.


Trainability is the capability of a model to be trained or fine-tuned to perform specific tasks. The video mentions that the API allows for more trainability, enabling users to shape the AI's behavior to better suit their needs.

💡System Prompt

A system prompt is a predefined set of instructions or cues given to an AI model to guide its responses. The video discusses the lack of control users have over the system prompt when using ChatGPT directly, versus the control provided by the API.

💡Chat Context

Chat context refers to the ongoing conversation history that an AI model uses to inform its responses. The video explains that the API allows users to control the chat context, selecting which parts of the conversation history to include in the AI's processing.


Insomnia is a REST client used for testing and developing APIs. In the video, it is mentioned as a tool the speaker uses to make API requests to OpenAI, similar to Postman, and is used to demonstrate how to interact with the OpenAI API.

💡Bearer Token

A bearer token is a security token that is used to access an API. The video script describes how to set the OpenAI token as a bearer token in Insomnia to authenticate API requests and receive responses from the AI model.


JSON (JavaScript Object Notation) is a lightweight data interchange format that is used to transmit data in web applications. The video mentions JSON as the format for the data sent in API requests to OpenAI, allowing users to specify parameters such as the system prompt and chat context.


The OpenAI API offers more flexibility and programmability than using ChatGPT directly.

API users have the ability to create customized versions of ChatGPT, like a personal trainer with a unique personality.

The API allows for more consistent and trainable interactions compared to the standard ChatGPT interface.

When using ChatGPT, the full chat history is sent to the OpenAI API, but users have no control over the system prompt.

API users can specify the model and version for more consistent results.

API requests can include a custom system prompt to guide the AI's behavior.

Users have control over the chat context sent to the API, allowing for more precise interactions.

The API provides a way to manage the chat history effectively, avoiding unwanted context.

API users can control the entire conversation flow, including starting a new session or modifying the prompt.

Using tools like Insomnia or Postman, users can easily make API requests to OpenAI.

API responses are more controlled and predictable, allowing for better integration into workflows.

The API enables users to inject specific instructions into the conversation, influencing the AI's responses.

Documentation and community resources are available to help users get started with the API.

The API provides a more powerful and customizable experience for users looking to leverage AI models.

Users can obtain an OpenAI token from ChatGPT to start making API requests.

Engaging with the API can lead to a better understanding and utilization of AI models.