Hi all!
With the announcement today has anyone got GPT Vision / DALLE 3 working through the API?
I am struggling to find how to ‘upload’ an image for Vision
DALLE 3 just produces errors
Keen to try TTS too if anyone has ?
Thanks!
Hi all!
With the announcement today has anyone got GPT Vision / DALLE 3 working through the API?
I am struggling to find how to ‘upload’ an image for Vision
DALLE 3 just produces errors
Keen to try TTS too if anyone has ?
Thanks!
Hey @KyleBehrend,
the OpenAI App does not have a specific module for this yet but you can use the Make an API Call Module.
{
"model": "gpt-4-vision-preview",
"messages": [
{
"role": "user",
"content": [
{
"type": "text",
"text": "PROMPT TEXT"
},
{
"type": "image_url",
"image_url": {
"url": "IMG_URL"
}
}
]
}
],
"max_tokens":3000
}
I can confirm that @Richard_Johannes approach works. Now I’m trying to get this to work with images sent via email. I’ve tried saving the attachment to OneDrive but GPT Vision doesn’t like loading from there. So now I’m trying to base64 encode the image before sending it to GPT Vision.
Joel,
Would you mind explaining how do I do to apply this information on make API call module? sorry if the question sounds stupid but, I dont know what to put on fields like, URL, method, headers, key value.
I’ve been trying to do it but it’s been hours with no progression
Thanks so much for the reply! Can you please help me, what do I populate here? I keep getting errors:
{
“model”: “dall-e-3”,
“prompt”: " TYPE YOUR PROMPT HERE",
“n”: 1,
“size”: “1024x1024”,
“quality”: “standard”,
“style”: “vivid”
}
Works great!! Thanks.
Thanks everyone! I cannot get a file from Google Drive - anyone figured that out yet?
im getting this on the code from
@Gian_Giannotti
[400] Your request contained invalid JSON: Extra data: line 10 column 1 (char 469)
{
“model”: “dall-e-3”,
“prompt”: "Create a cinematic and visually striking image that embodies the world of cryptocurrency and crypto trading. Imagine screens displaying real-time crypto trading stats, fluctuating graphs, and various cryptocurrency logos like Bitcoin, Ethereum, and others, symbolizing the global interconnectedness of crypto trading. DO NOT ADD TEXT OR CHARACTERS ",
“n”: 1,
“size”: “1024x1024”,
“quality”: “standard”,
“style”: “vivid”
}
AuthenticationFailed
Server failed to authenticate the request. Make sure the value of Authorization header is formed correctly including the signature. RequestId:cdb5f612-701e-0024-66f6-11583e000000 Time:2023-11-08T03:52:29.1361141Z
Signature fields not well formed.
this is the result?
I think the API is down atm https://status.openai.com/
Try not to use dots in your prompt as this causes error
quotation marks neither
[400] Your request contained invalid JSON: Expecting property name enclosed in double quotes: line 2 column 1 (char 2)
this would be great, there is even the option now to post file or image URL, but it still won’t accept.
is there an explanation for this already somewhere?
Attached here is an installation for the models I developed:
https://www.make.com/en/hq/app-invitation/91cd0339d2ab1e9a7ca070171112e284
Hi Rabin
Thank you for doing this
Could you tell me how do I do to download to text to speech file. I can only see a binary result, with no link to download.
Welcome to the Make community!
Your questions seem to have strayed from the original question that the thread starter initially asked.
In future, please create a new thread for each question. This makes it easier for others with the same problem to search for the answer, and for you to get the answers you need, since new threads have a different notification channel, and solved threads may be ignored by others. Thank you for your cooperation!
Sorry about that
I was so surprised with the module he created that I forgot the main purpose