Our dev platform team is planning their work for the next month, and wants to make sure we're building what people want. So if you're an OpenAI API user — what should we add/fix/improve to make your experience better?

Mar 8, 2023 · 5:10 PM UTC

353
54
7
521
Load newest
Replying to @gdb
Larger output limit
Replying to @gdb
More tokens and/or session for ChatGPT. Streaming mode for whisper
Replying to @gdb
For summarization, the api should accept documents beyond the attention window and automatically break them into logical blocks that do fit in the attention window, concatenation the resulting summaries and returning it. Cohere supports this which is why we switched for some uses
Replying to @gdb
Please, fine-tune for chatgpt model, not only base models. Thanks
Replying to @gdb
Drag and drop folder system for chats or self organising chats for ChatGPT
Replying to @gdb
A better API key management pls for better tracking.
Replying to @gdb
Would be great to have more information on the existing models for research purposes. Info like size of the models, training methods, etc. There are currently some info on the website but not for all models available in the API.
Replying to @gdb
Hitting rate limit (max 20/min) with each call being separated by time.sleep(10) 🧐🧐🧐
Replying to @gdb
The option to output code with less description. More direct approach.
Replying to @gdb
+1 for the context
Load more