Vladimir

Like in #10 generation id needs to be cached the endpoint is `https://labs.openai.com/api/labs/collections/private/generations` ```python import requests auth_header = "Bearer sess-......" payload = {"generation_ids": ["generation-ExyZNQNG5xzIkYqmPfOejUzP"]} ret = requests.post( f"https://labs.openai.com/api/labs/collections/private/generations", headers={"Authorization": auth_header},...

Off topic: I'm actually quite shocked, that no one has done what I just did. This library can be (and probably will be) very helpful for developers before OpenAI releases...

@ezzcodeezzlife do you need PR for any of these?

For the text models (GPT-3) OpenAI released well-documented API and you can get permanent API key for yourself. I assume there will be same situation once they release DALLE-2 to...

So, there's additional steps required. After generation of images it returns generation with id ```json { "generations": { "data": [ { "id": "generation-KdiiXg780fGKdfOEiMXI8Tnf", ... }, .., ] } } ```...

```python import requests auth_header = "Bearer sess-......." ret = requests.get( "https://labs.openai.com/api/labs/billing/credit_summary", headers={"Authorization": auth_header}, ) print(ret.json()) ``` returns `{'aggregate_credits': 545, 'next_grant_ts': 1661754850, 'breakdown': {'free': 0, 'paid_dalle_15_115': 545}, 'object': 'credit_summary'}`

So, endpoint is `https://labs.openai.com/api/labs/tasks`, but payload is different. Pay attention to batch_size - it's limited to 3. I'm not sure what is the reasoning behind that (aside their current UI...

That's for the already generated image. About uploads I will have to look later, they do base64 encoding, but i'm not sure they do it the way I do it....