I have some questions about Chat GPT Pricing
Token Limits:
What is the maximum number of tokens allowed per API request or per minute/hour/day?
Are there different token limits for different models (e.g., GPT-3.5 vs. GPT-4)?
How are tokens counted (e.g., by characters, words, subwords)?
Are there overage charges for exceeding token limits?
Pricing Structure:
What is the cost per token for different models?
Are there different pricing tiers based on usage volume?
Are there any fixed monthly fees or free tiers?
Are there discounts for long-term commitments or high-volume usage?
API vs. Pro Versions:
What are the key differences between the API version and the Pro version in terms of features, performance, and support?
Can I use my own API key with the Pro version? If so, will I be charged by both the LLM provider and my API provider?
Data Handling:
Can I upload large text files (e.g., 100,000+ characters) for analysis?
Are there limits on file size or format?
What kind of analyses can I perform on uploaded data (e.g., summarization, sentiment analysis, pattern recognition)?
How secure is my data when uploaded to the platform?
Fine Print:
What are the terms of service and usage policies?
Are there any hidden fees or limitations not explicitly stated in the pricing information?
What happens if I exceed usage limits or violate the terms of service?
Welcome to the Community! A few links to get you started.
1. General overview of the models including specific technical details
https://openai.com/api/
https://platform.openai.com/docs/models/overview
2. Model pricing
https://openai.com/api/pricing/
3. Applicable limits
Limits for the use of different models is determined based on the Tier you are in. The link provided shows the different tiers including the applicable limits and the criteria that must be met.
https://platform.openai.com/docs/guides/rate-limits/usage-tiers
4. Key policies
https://openai.com/policies/
Finally, please note that there is no pro version or some such for the API. Your access to different models and the applicable limits is entirely determined by the Tier you are in, which in turn is determined by the amount you have spent on API consumption.
It is also worth noting that there is no longer a free tier, meaning OpenAI does no longer supply free API credits at the point when a new developer account is created.
1 Like
My questions are specific. Those are stock answers which address 20% of my questions.
I do believe that 80% of your questions can be addressed with the information under these links. Have you had a look?
I’ll give the short answer to those that require some in-between reading the lines:
- Are there overage charges for exceeding token limits?
Once you exceed pre-defined limits you simply get an error message.
Are there different pricing tiers based on usage volume?
No.
Are there any fixed monthly fees or free tiers?
No.
Are there discounts for long-term commitments or high-volume usage?
No. Unless you can negotiate with the OpenAI sales team. No official info available in this regard.
What are the key differences between the API version and the Pro version in terms of features, performance, and support?
No Pro version available as previously indicated.
Can I use my own API key with the Pro version? If so, will I be charged by both the LLM provider and my API provider?
Not applicable.
Can I upload large text files (e.g., 100,000+ characters) for analysis?
With gpt-4-turbo models you can, either in the form of adding the content from the files to the context or by using tools like the Assistant.
What kind of analyses can I perform on uploaded data (e.g., summarization, sentiment analysis, pattern recognition)?
The sky is the limit.
How secure is my data when uploaded to the platform?
https://openai.com/enterprise-privacy/
What happens if I exceed usage limits or violate the terms of service?
You’ll be banned for violations. For exceeding usage there is no immediate impact other than not being to make API calls for a defined period of time.
1 Like