OpenAI LLMs

Creator of the GPT family and the ChatGPT product. Pioneer of large-scale RLHF and reasoning-trained models.

OpenAI ships frontier models behind a unified Chat Completions / Responses API. Its strongest commercial advantage is the breadth of its model menu — GPT-4o for general tasks, GPT-4.1 for long-context coding, the o-series for reasoning, and embedding/Whisper/TTS endpoints alongside. Token costs vary widely across this menu, which is why a per-call calculator matters more here than for single-model providers.

Founded: 2015 · HQ: San Francisco, USA · Docs: platform.openai.com ↗

All OpenAI models

ModelFamilyContextInput $/MOutput $/MReleasedStatus
GPT-5 GPT-5 400K $1.25 $10.00 2025-08-07 active
GPT-5 Mini GPT-5 400K $0.25 $2.00 2025-08-07 active
GPT-5 Nano GPT-5 400K $0.05 $0.40 2025-08-07 active
o3 o-series 200K $2.00 $8.00 2025-04-16 active
GPT-4.1 GPT-4 1.0M $2.00 $8.00 2025-04-14 active
o3-mini o-series 200K $1.10 $4.40 2025-01-31 active
GPT-4o Mini GPT-4 128K $0.15 $0.60 2024-07-18 active
GPT-4o GPT-4 128K $2.50 $10.00 2024-05-13 active

Comparisons with other providers

The most-searched comparisons involving OpenAI models:

Working with the OpenAI API

Documentation lives at https://platform.openai.com/docs. Before paying for any call, count input tokens with the appropriate counter: