Chat
OpenAI models
OpenAI models are a series of powerful language models developed by OpenAI.
Using the API
Use Chat completion API to call OpenAI models. See the pricing page for pricing details.OpenAI models comparison
| Feature | GPT-5 | GPT-4.1 | GPT-4.1-mini | GPT-4.1-nano | GPT-4o | GPT-4o-mini | o4-mini |
|---|---|---|---|---|---|---|---|
| API model key | gpt-4.1 | gpt-4.1 | gpt-4.1-mini | gpt-4.1-nano | gpt-4o | gpt-4o-mini | o4-mini |
| Description | Latest GPT-5 model | Mainstream GPT-4.1 | Lightweight GPT-4.1 | Ultra-light GPT-4.1 | Latest GPT-4o model | Lightweight GPT-4o | Cost-effective GPT-o |
| Current version | gpt-5 | gpt-4.1 | gpt-4.1-mini-2024-07-18 | gpt-4.1-nano-2024-07-18 | gpt-4o-2024-11-20 | gpt-4o-mini-2024-07-18 | o4-mini |
| Best for | Advanced reasoning and complex problem-solving | General reasoning | Efficient general-purpose tasks | Fast, low-cost tasks | Complex reasoning and creative tasks | Efficient general-purpose tasks | Budget-friendly tasks |
| Vision capability | Yes (Enhanced) | No | No | No | Yes | Yes (see note) | No |
| Tool calls | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Context window | 400k tokens | 128K tokens | 128K tokens | 128K tokens | 128K tokens | 128K tokens | 128K tokens |
| Max output | 32k tokens | 8K tokens | 8K tokens | 8K tokens | 16K tokens | 16K tokens | 8K tokens |
| Knowledge cutoff | Jun 2024 | Oct 2023 | Oct 2023 | Oct 2023 | Oct 2023 | Oct 2023 | Oct 2023 |
Important notes
Vision capability
- Both GPT-4o and GPT-4o-mini support vision capabilities, allowing them to process and analyze images
- ⚠️ Known issue with GPT-4o-mini: There are tokenization issues when processing images with GPT-4o-mini that may result in unexpectedly high input token counts. For vision tasks at this price point, we recommend using Claude 3 Haiku instead.
Model versions and aliases
GPT-4.1
gpt-4.1- Mainstream GPT-4.1 modelgpt-4.1-mini- Points to the current recommended version (currently gpt-4.1-mini-2024-07-18)gpt-4.1-mini-2024-07-18- Specific versiongpt-4.1-nano- Points to the current recommended version (currently gpt-4.1-nano-2024-07-18)gpt-4.1-nano-2024-07-18- Specific version
GPT-4o
gpt-4o- Points to the current recommended version (currently gpt-4o-2024-11-20)gpt-4o-2024-11-20- Specific version
GPT-4o-mini
gpt-4o-mini- Points to the current recommended version (currently gpt-4o-mini-2024-07-18)gpt-4o-mini-2024-07-18- Specific version
o4-mini & o4 series models
o4-mini- Cost-effective GPT-o model- o4 series models (such as o4-mini) include reasoning steps in their responses. This means they may generate more detailed explanations and therefore consume more output tokens than other models. Please consider this when planning your token budget.
Official documentation
- Vision Guide - Learn more about using vision capabilities