When hitting the job limit, try estimating when there will be a slot again

[johny-b]

So, when the error is

openai.RateLimitError: Error code: 429 - {'error': {'message': "This fine-tune request has been rate-limited. Your organization has reached the maximum of [YOUR NUMBER] fine-tuning requests per day for the model 'gpt-4.1-2025-04-14'.", 'type': 'invalid_request_error', 'param': None, 'code': 'daily_rate_limit_exceeded'}}

See the number of FT jobs and when they were created.
Note that we don't have FT jobs for other projects. So this will only work reliably if all jobs are in the same project.

So, it could say e.g.:

• Your limit is 8, your project has 8 within last 17 hours, so expect a free slot in 7 hours
• Your limit is 8, your project has only 5 within the last 5 hours, so surely there are jobs in other projects. Regardless, this project will free a slot in 3h, but maybe some other will free one sooner.

[johny-b]

hopefully resolved in 9b417fd