Procesamiento prioritario para clientes de la API
Fast mode offers reliable, high-speed performance with the flexibility to pay-as-you-go. For our latest frontier model, gpt-5.6-sol, you can access up to 2.5x faster speeds with Fast mode.
By choosing Fast mode, you can unlock:
- Predictably low latency: Fast mode generates tokens faster and at a more consistent speed than the Standard processing service, even during peak demand.
- Easy-to-use flexibility: Like Standard processing, Fast mode can be accessed on a flexible, pay-as-you-go basis instead of requiring advance provisioning.
Note: Priority processing was renamed Fast mode on July 30, 2026. You can use either service_tier: priority or service_tier: fast in your API requests.
| Precio por 1 millón de tokens de entrada | Precio por 1 millón de tokens de entrada (en caché) | Precio por 1 millón de tokens de salida | Tiempo en funcionamiento garantizado3 | Latencia garantizada3 | |
|---|---|---|---|---|---|
GPT-5.6 Sol excluye el contexto extenso1 | USD 10.00 | USD 1.00 | USD 60.00 | 99.9% | 99% > 80 tokens por segundo2 |
GPT-5.6 Terra excluye el contexto extenso1 | USD 4.00 | USD 0.40 | USD 24.00 | 99.9% | 99% > 70 tokens por segundo2 |
GPT-5.6 Luna excluye el contexto extenso1 | USD 0.40 | USD 0.04 | USD 2.40 | 99.9% | 99% > 100 tokens por segundo2 |
GPT-5.5 excluye el contexto extenso1 | USD 12.50 | USD 1.250 | USD 75.00 | 99.9% | 99% > 50 tokens por segundo2 |
GPT-5.4 mini excluye el contexto extenso1 | USD 1.50 | USD 0.150 | USD 9.00 | 99.9% | 99% > 100 tokens por segundo2 |
GPT-5.4 excluye el contexto extenso1 | USD 5.00 | USD 0.500 | USD 30.00 | 99.9% | 99% > 50 tokens por segundo2 |
GPT-5.2 | USD 3.50 | USD 0.350 | USD 28.00 | 99.9% | 99% > 50 tokens por segundo2 |
GPT-5.1 | USD 2.50 | USD 0.250 | USD 20.00 | 99.9% | 99% > 50 tokens por segundo2 |
GPT-5 | USD 2.50 | USD 0.250 | USD 20.00 | 99.9% | 99% > 50 tokens por segundo2 |
GPT-5 mini | USD 0.45 | USD 0.045 | USD 3.60 | 99.9% | 99% > 80 tokens por segundo2 |
GPT-5.1 codex | USD 2.50 | USD 0.250 | USD 20.00 | 99.9% | 99% > 50 tokens por segundo2 |
GPT-5 codex | USD 2.50 | USD 0.250 | USD 20.00 | 99.9% | 99% > 50 tokens por segundo2 |
GPT-4.1 | USD 3.50 | USD 0.875 | USD 14.00 | 99.9% | 99% > 80 tokens por segundo2 |
GPT-4.1 mini | USD 0.70 | USD 0.175 | USD 2.80 | 99.9% | 99% > 90 tokens por segundo2 |
GPT-4.1 nano | USD 0.20 | USD 0.050 | USD 0.80 | 99.9% | 99% > 100 tokens por segundo2 |
GPT-4o gpt-4o-2024-11-20 gpt-4o-2024-08-06 | USD 4.25 | USD 2.125 | USD 17.00 | 99.9% | 99% > 80 tokens por segundo2 |
gpt-4o-2024-05-13 | USD 8.75 | — | USD 26.25 | 99.9% | 99% > 80 tokens por segundo2 |
GPT-4o mini | USD 0.25 | USD 0.125 | USD 1.00 | 99.9% | 99% > 90 tokens por segundo2 |
o3 | USD 3.50 | USD 0.875 | USD 14.00 | 99.9% | 99% > 80 tokens por segundo2 |
o4-mini | USD 2.00 | USD 0.500 | USD 8.00 | 99.9% | 99% > 90 tokens por segundo2 |
¿Cómo funciona?
Customers can direct traffic to Fast mode on a per request basis using the existing service_tier parameter, with the option service_tier = "fast".
Tokens served by Fast mode will be billed on a per-token basis, priced at a premium relative to Standard processing rates.
In addition to being configured at the request level, you can also default a project to Fast mode in Project settings > Default Service Tier: Fast. You can still override per request. Selecting Fast in your project settings is equivalent to selecting Priority.
Limitaciones
- Fast mode rate limits are shared with other service tiers.
- In rare cases, rapid increases to your Fast mode Tokens per Minute can lead to hitting ramp rate limits. If you exceed the ramp rate limit, then additional traffic may be sent to Standard processing instead.