Fix OpenAI 404 Model Not Found

Fix OpenAI 404 Model Not Found Error

The 404 Not Found error with the message "The model does not exist or you do not have access to it" is a common issue when working with the OpenAI API. This error means either the model name is incorrect, the model has been deprecated, or your account does not have access to it. Here is how to diagnose and fix every variant of this error.

Understanding the Error

{
  "error": {
    "message": "The model `gpt-4o-latest` does not exist or you do not have access to it.",
    "type": "invalid_request_error",
    "param": null,
    "code": "model_not_found"
  }
}

Cause 1: Typo in Model Name

Model names must be exact. Here are common mistakes:

• gpt-4-o should be gpt-4o (no hyphen before "o")
• gpt4o should be gpt-4o (missing hyphen)
• GPT-4o should be gpt-4o (lowercase only)
• gpt-4o-latest may not be a valid snapshot — use gpt-4o
• gpt-35-turbo should be gpt-3.5-turbo (dot, not hyphen)
• dall-e-3 is correct for image generation, but cannot be used with chat completions

Valid Model Names (2026)

Here is a reference list of currently available models:

# GPT-4o family
gpt-4o
gpt-4o-mini

# GPT-4 family
gpt-4-turbo
gpt-4

# GPT-3.5 family
gpt-3.5-turbo

# Reasoning models
o1
o1-mini
o3-mini

# Embedding models
text-embedding-3-large
text-embedding-3-small
text-embedding-ada-002

# Image generation
dall-e-3
dall-e-2

# Text-to-speech
tts-1
tts-1-hd

# Speech-to-text
whisper-1

Cause 2: Deprecated or Retired Model

OpenAI periodically retires older model versions. If you are using a dated snapshot, it may no longer be available:

• gpt-4-0314 — Retired, use gpt-4
• gpt-4-0613 — Retired, use gpt-4 or gpt-4-turbo
• gpt-3.5-turbo-0301 — Retired, use gpt-3.5-turbo
• text-davinci-003 — Retired, use gpt-3.5-turbo
• code-davinci-002 — Retired, use gpt-4o

Cause 3: Account Access Restrictions

Some models require specific account tiers or approval:

• Free tier accounts may not have access to all models
• GPT-4 requires at least a Tier 1 account (paid)
• o1 and reasoning models may require higher usage tiers
• Fine-tuned models are only accessible to the organization that created them

Cause 4: Wrong API Endpoint

Using the wrong endpoint for a model type will return 404:

# Wrong - DALL-E with chat completions endpoint
POST /v1/chat/completions
{"model": "dall-e-3", ...}  # 404!

# Correct - DALL-E with images endpoint
POST /v1/images/generations
{"model": "dall-e-3", "prompt": "A sunset over mountains"}

# Wrong - Embedding model with chat endpoint
POST /v1/chat/completions
{"model": "text-embedding-3-small", ...}  # 404!

# Correct - Embedding with embeddings endpoint
POST /v1/embeddings
{"model": "text-embedding-3-small", "input": "Hello world"}

How to List Available Models

from openai import OpenAI

client = OpenAI(base_url="https://claude4u.com/v1")

# List all models your account can access
models = client.models.list()
for model in sorted(models.data, key=lambda m: m.id):
    print(model.id)

// Node.js
import OpenAI from 'openai';

const client = new OpenAI({ baseURL: 'https://claude4u.com/v1' });
const models = await client.models.list();
models.data
    .sort((a, b) => a.id.localeCompare(b.id))
    .forEach((m) => console.log(m.id));

# cURL
curl https://claude4u.com/v1/models \
  -H "Authorization: Bearer $OPENAI_API_KEY"

Cause 5: Organization or Project Mismatch

If you belong to multiple organizations, a model available in one org may not be in another:

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-key",
    organization="org-correct-org-id",
    base_url="https://claude4u.com/v1"
)

Quick Debugging Steps

1. Double-check the exact model name for typos and case
2. List available models with the /v1/models endpoint
3. Check if the model has been deprecated on OpenAI's documentation
4. Verify your account tier has access to the requested model
5. Ensure you are using the correct API endpoint for the model type
6. Try using the base model name without a date snapshot suffix