GPT-4o
OpenAIMULTIMODAL Proprietary
- Provider
- OpenAI
- Modality
- MULTIMODAL
- Parameters
- —
- Context window
- 128,000 tokens
- Weights
- Proprietary
- Released
- 13 May 2024
GPT-4o (May 2024) is the original multimodal workhorse, now legacy — kept mainly for vision/voice pipelines that newer cheap text models can't cover.
Best for
- Legacy multimodal apps needing text + vision (+ audio) input
- Existing integrations not yet migrated to GPT-5
- Real-time / voice features historically built on the 4o stack
How it compares — and the India angle
- Now a legacy model — for new multimodal work use a GPT-5.x model instead.
- Pricier output and smaller 128K context than GPT-4.1 and the GPT-5 line — rarely the economical pick today.
- If you only need text, GPT-4.1 or GPT-5.4 mini beats it on price and quality.
Benchmarks
Representative public scores (approximate, higher is better) for relative comparison. Check the provider for the latest official results.
MMLU89
How to access
API ~$2.50 / 1M input, ~$10 / 1M output. API-only legacy; full text+vision+audio input.
Access GPT-4o