Model pricing
GPT-4o Realtime pricing
GPT-4o Realtime by OpenAI costs $5 per 1M input tokens and $20 per 1M output tokens (cached input $2.5, batch input not recorded) as of 2026-10-31. A 1,000-input/500-output request costs $0.015. The lowest-cost alternative with at least this context window and the same recorded reasoning category is Ling 2.6 Flash at $0.01/$0.03; capability equivalence is not guaranteed.
Notes
Low-latency speech-to-speech Realtime API model
Key pricing facts
| Fact | Value |
|---|---|
| Input / 1M tokens | $5 |
| Output / 1M tokens | $20 |
| Cached input / 1M | $2.5 |
| Batch input / output / 1M | Not recorded |
| Context window | 128,000 |
| Provider | OpenAI |
| Last verified | No model-specific verification date recorded |
| Snapshot updated | 2026-10-31 |
Last updated: · Source: OpenAI pricing · Methodology
Frequently asked questions
What does GPT-4o Realtime cost per million tokens?
Input costs $5 and output costs $20 per million tokens. The 1,000-input/500-output example costs $0.015, before discounts, taxes, hidden reasoning tokens or context-tier surcharges.
Does GPT-4o Realtime have cached or batch pricing?
Cached input: $2.5 per million tokens under the existing provider cache rules. Batch: not recorded in this dataset. Eligibility and cache-write charges depend on the provider; these discounts are not automatically stacked.
What is a cheaper alternative to GPT-4o Realtime?
Ling 2.6 Flash costs $0.01 input and $0.03 output per million tokens. This comparison matches context capacity and reasoning category, not benchmark quality. Test your workload before replacing a model.
How often are these prices verified?
The live pricing table synchronizes stored model records daily; public file snapshots update when the site is rebuilt. Synchronization is not independent provider verification. Snapshot updated 2026-10-31. No model-specific verification date is recorded; the snapshot date must not be interpreted as an official provider verification.