Hi, Im building a project multichat via whastapp with assistants API key. Now, when I have it 99% finished, rate limits stop me. I was using gpt-4.1, it was good, it followed prompt instructions, vector stores, tools functions, etc. Now, I have to use 4.1-mini, but, with this, its impossible to use a intelligent model in a whatsapp conversation. Leaving aside 4th or 4.1 and other core models, the rest of the lightweight models are really stupid and useless for assistants. So, is there a way to raise the rate limit on base models (4th, 4.1, etc.) without having to waste $100 just to move up a tier? If I explain my project to OpenAI, will they raise the limit? Does anyone know ways to raise the rate limit? It’s urgent, thanks.
No, there is nothing you can do to bypass the rules. I was having a similar issue. Had a great idea, tried to build it, ran into the token limit. I asked support and they replied there is no way to upgrade early. ChatGPT told me to burn some credits, by letting it do pointless things, but that obviously is just as stupid. I haven’t had much issues after simplyifing my system prompt so far. Or ChatGPT told me to specify a rate limit per message/run, but I haven’t tried that yet. Totally agree with the headline, though…