OpenAI
ChatGPT has a partial response
OpenAI
Interact with ChatGPT and generate images with DALL·E through flows.
This app lets you talk directly with OpenAI ChatGPT through flows.
After installing the app, set up a free API key in the app settings. The API key will work for up to 3 months before it lapses. After the trial period the API key can be reactivated with a subscription.
Please follow the following guidelines when creating flows, or your account will be in danger of being shut down:
The following use-cases are prohibited:
* Illegal or harmful industries
* Misuse of personal data
* Promoting dishonesty
* Deceiving or manipulating users
* Trying to influence politics
The following types of content is not allowed:
* Hate
* Harassment
* Violence
* Self-harm
* Sexual
* Political
* Spam
* Deception
* Malware
Ondersteunde apparaten
Flow kaarten
Klik op een apparaat hierboven om gerelateerde Flow kaarten te zien.
Als...
When ChatGPT is answering a question the first partial answer will trigger this card. The proceeding partial answers will be delayed until the flow action 'Tell ChatGPT to send next partial response' is called. This is to ensure that the text-to-speech engine has time to finish processing a partial answer before a new one arrives.
OpenAI
ChatGPT has finished answering
This card will only trigger when all the partial answers have been received. This could take some time if you ask for something like a fairytale.
Warning: The response string may be too long to be accepted by a text-to-voice assistant.
OpenAI
A webhook triggered with a flag and a message
Use webhooks to allow for other input sources to this app. See the app settings help section to see what the webhook address is. If you add a flag then you can for example use it to choose whether or not to forward the text to image generation or to ChatGPT
Dan...
ChatBot
Send the question ... to the chatbot and return the answer
Advanced
ChatBot
Clear chat history
CompletionBot
Send prompt ... to completion bot and return the answer
Advanced
OpenAI
Ask ChatGPT the question Question to ask ChatGPT and _WAIT_ for the full answer
This is a very slow operation and should ONLY be used if you don't need a quick response, also the answer is not appropriate for text-to-speech clients who can't handle responses longer than ~200 characters. For this, you should use partial responses instead.
The question provided with this flow card will be sent to ChatGPT for processing. When completed, the answer will be returned in the text Token {{ChatGPT_FullResponse}}.
Advanced
OpenAI
Ask ChatGPT the question Question to ask ChatGPT and trigger partial responses
The question provided with this flow card will be sent to ChatGPT for processing. A trigger card is then triggered once the first partial answer arrive. Then when you have finished processing the partial response, you need to tell ChatGPT to send the next partial answer to trigger the next partial answer.
OpenAI
Tell ChatGPT to send next partial response
The pending partial responses are delayed until requested by this action. This is to prevent the text-to-speech device to be overwhelmed with commands.
OpenAI
Generate an image of size Image Size from the text Description and return it
Uses DALL·E to generate images according to your description.
Advanced
Ondersteuning
Heb je een probleem met deze app? Neem contact op met de ontwikkelaar via het Homey Community Forum.
Wat is nieuw
Versie 1.6.3 — Added GPT-4o and GPT-4-turbo support + increased max_token length for main queries.
OpenAI
OpenAI is niet compatibel met de geselecteerde Homey
OpenAI wordt op Homey geïnstalleerd.
Changelog
We sturen een bedankje naar Frode Heggelund!
We laten Frode Heggelund weten dat je dol bent op de app. Overweeg om Frode Heggelund te ondersteunen door een donatie te maken?
Doneer aan Frode Heggelund
De ontwikkelaar van OpenAI accepteert donaties.
Support de ontwikkelaar en doe een donatie!
Je voorstel is verzonden
We laten Frode Heggelund weten dat je een voorstel hebt verzonden.
