Feature request for product/service
AI Models
Describe the request
I want to know which model I should use when my paid model quota is exhausted, and I still get a “quota exceeded” prompt even after switching to Auto mode with 2.0 Version.
I don’t think there are any built-in free models anymore in the current pricing system. See the current model pricing table in the docs.
According to the Auto pricing docs, Auto is currently priced like so:
- Input + Cache Write: $1.25 per 1M tokens
- Output: $6.00 per 1M tokens
- Cache Read: $0.25 per 1M tokens
I haven’t tried this myself, but you may be able to use one of the free models on OpenRouter or a similar provider. I’m sure there are up to date guides for setting this up out there. Just note that if you’re using external providers, Cursor’s privacy settings can’t apply. That’s explained in the API Keys doc page.
You could try GPT-5-mini (which does not add to my monthly quota, but maybe that is only before you meet your quota) or try some promotional free models like grok-code. Also try some DeepSeek models.
@deanrie I can see Gemini 2.5 flash consuming premium requests. This wasn’t the case around a week ago.
What changed? Is it a bug?
Also, please let me know which free models we can use? (I only know grok-code-fast-1)
@MidnightOak Even gpt-5-mini is not free anymore
1 Like
Only Grok Code , GPT-5 Nano are free now.
Deepseek-v3.1 costs 0.3 Premium Requests, it is not free any more.
I bought windsurf pro yesterday. These are free models in windsurf. Just several weeks ago, there were more free models in cursor than windsurf.
1 Like
That’s so sad, gpt5-mini is not free started from yesterday
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.