This error means the model name in your API request does not match any available model in your OpenAI account. Common causes include typos in the model name, using a model that requires a higher-tier API plan, or referencing a model that has been deprecated or renamed. Check the exact model ID in OpenAI's model list and update your code.
What does "The model `gpt-4o` does not exist" mean?
When the OpenAI API returns "The model `gpt-4o` does not exist," it means the model identifier you sent in the 'model' field of your API request does not match any model available to your account. The API's model catalog is extensive and changes frequently — models are added, renamed, and deprecated regularly. A model that worked last week may have been replaced or restricted.
This error is especially confusing because model naming conventions are inconsistent. Models like gpt-4o, gpt-4o-mini, gpt-4-turbo, and gpt-4 are all separate identifiers. The 'o' in gpt-4o stands for 'omni' and is a distinct model, not a typo. Using 'gpt4o' (without the hyphen) or 'GPT-4o' (wrong case) will trigger this error because model names are exact-match and case-sensitive.
Access restrictions add another layer. Some models are only available to accounts on specific API tiers. New accounts start with Tier 1 access and may not have access to the latest models until they accumulate sufficient usage. Enterprise and organization accounts may have different model availability than individual accounts.
Common causes
A typo in the model name
model identifiers are case-sensitive and require exact hyphens (gpt-4o not gpt4o)
The model requires a higher
API access tier than your account currently has
The model was deprecated or
replaced with a newer version and the old identifier no longer works
Your API key belongs to a different organization that
does not have access to this specific model
The model is in limited
preview/beta and not yet available to general API users
New reasoning models (o1, o3,
GPT-5 series) use different naming conventions than the GPT-4 family
How to fix the model-not-found error in OpenAI API
First, check the exact model ID you need. Go to the OpenAI platform documentation (platform.openai.com/docs/models) to see the current list of available models and their exact identifiers. Common current models include gpt-4o, gpt-4o-mini, gpt-4-turbo, and gpt-3.5-turbo. Verify the spelling, hyphens, and case match exactly.
If the model name is correct, check your account's API tier at platform.openai.com/account. Some models require Tier 2 or higher access, which requires a minimum spend history. You can also call the models list endpoint (GET /v1/models) with your API key to see every model available to your specific account. If you need a model that is not available on your tier, add funds to your account and make some API calls to build usage history. For projects that need specific model access, RapidDev can help configure the correct API settings and implement model fallback logic.
import openai# Wrong model name — typo or deprecatedresponse = openai.chat.completions.create( model="gpt4o", # Missing hyphen messages=[{"role": "user", "content": "Hello"}])import openai# Correct model nameresponse = openai.chat.completions.create( model="gpt-4o", # Correct with hyphen messages=[{"role": "user", "content": "Hello"}])# Or list available models to check accessmodels = openai.models.list()for model in models.data: if 'gpt-4' in model.id: print(model.id)Prevention tips
- Always copy model names from the official OpenAI documentation — do not type them from memory, as hyphens and case matter
- Call the /v1/models endpoint to see exactly which models your API key has access to
- Implement model fallback logic — if gpt-4o is unavailable, fall back to gpt-4o-mini or gpt-3.5-turbo
- Check OpenAI's deprecation schedule — deprecated models stop working on their sunset date with no grace period
Still stuck?
Copy one of these prompts to get a personalized, step-by-step explanation.
The OpenAI API returns 'The model gpt-4o does not exist' but I can use it in ChatGPT. How do I check which models my API key has access to and what tier I need?
Update my OpenAI API code to use the correct model name and add fallback logic for when a model is unavailable.
Frequently asked questions
Why does OpenAI say "The model `gpt-4o` does not exist" when I can use it in ChatGPT?
ChatGPT and the API are different products with different access controls. Your API account may be on a lower tier that does not include gpt-4o. Check your account tier at platform.openai.com/account.
Are OpenAI model names case-sensitive?
Yes. Model names must be lowercase with exact hyphens. 'GPT-4o', 'gpt4o', and 'gpt-4O' are all invalid. The correct name is 'gpt-4o'.
How do I check which models my OpenAI API key can access?
Call the models list endpoint: GET https://api.openai.com/v1/models with your API key. It returns every model your key has access to. You can also use the Python SDK: openai.models.list().
What happens when an OpenAI model is deprecated?
Deprecated models stop working on their published sunset date. API calls with the old model name return a 'model not found' error. OpenAI provides a deprecation schedule and recommends migration paths to newer models.
Do new reasoning models (o1, o3) work with the same code as GPT-4?
Not entirely. Reasoning models reject the temperature parameter (only default value 1 is allowed) and use reasoning.effort instead. They also have different pricing and may require different API tier access.
Talk to an Expert
Our team has built 600+ apps. Get personalized help with your issue.
Book a free consultation