OpenAI models on One API
25 models we could call on 22 November 2026, from $0.02 per million input tokens.
OpenAI in numbers
- Models we could call
- 25
- our probe of 22 November 2026 · measured by us
- Median time to first token
- 1,047 ms
- our probe of 22 November 2026 · measured by us
Every OpenAI model
| Model | Input, $ per 1M | Output, $ per 1M | Context, tokens | First token | Tokens a second |
|---|---|---|---|---|---|
| text-embedding-3-small | $0.02 | n/a | 8,191 | 374 ms | n/a |
| gpt-oss-120b | $0.0463 | $0.2125 | 131,072 | 1,060 ms | 11.2 |
| gpt-5-nano | $0.05 | $0.4 | 400,000 | 2,712 ms | 235.4 |
| gpt-6-luna | $0.1 | $0.5 | 1,050,000 | 995 ms | 179.1 |
| text-embedding-3-large | $0.13 | n/a | 8,191 | 455 ms | n/a |
| gpt-4o-mini | $0.15 | $0.6 | 128,000 | 827 ms | 72.1 |
| gpt-5.4-nano | $0.2 | $1.25 | 400,000 | 451 ms | 159.3 |
| gpt-5.6-luna | $0.2 | $1.2 | 1,000,000 | 797 ms | 220 |
| gpt-5-mini | $0.25 | $2 | 400,000 | 4,164 ms | 124.6 |
| gpt-4.1-mini | $0.4 | $1.6 | 1,000,000 | 857 ms | 132.1 |
| gpt-5.4-mini | $0.75 | $4.5 | 400,000 | 575 ms | 269.3 |
| gpt-6.1-terra | $1 | $5 | 1,050,000 | 1,134 ms | 94.8 |
| gpt-5 | $1.25 | $10 | 400,000 | 4,725 ms | 167.3 |
| gpt-5-search-api | $1.25 | $10 | 400,000 | 1,232 ms | 234.5 |
| gpt-5.2 | $1.75 | $14 | 400,000 | 597 ms | 120.6 |
| gpt-4.1 | $2 | $8 | 1,000,000 | 551 ms | 296.5 |
| gpt-5.6-terra | $2 | $12 | 1,000,000 | 845 ms | 222.1 |
| gpt-6.1-sol | $2 | $10 | 1,050,000 | 1,298 ms | 183.2 |
| o3 | $2 | $8 | 200,000 | 2,050 ms | 229.2 |
| gpt-4o | $2.5 | $10 | 128,000 | 506 ms | 224.5 |
| gpt-5.4 | $2.5 | $15 | 400,000 | 757 ms | 175.6 |
| gpt-5.6-sol | $4 | $20 | 1,000,000 | 1,047 ms | 245.3 |
| gpt-5.5 | $5 | $30 | 400,000 | 1,328 ms | 122.5 |
| gpt-6.2-astra | $8 | $40 | 1,050,000 | 1,292 ms | 94.5 |
| gpt-6-astra | $10 | $50 | 1,050,000 | 1,408 ms | 186.7 |
OpenAI's own price list
- console.groq.comThe source of the list prices on this page
Other providers
How this page is made
Prices, context sizes and abilities are read from One API's public price feed and its model endpoints each time this page is built. The time to the first token, the output speed and the price of a call are measured by us: 3 calls for each model through a One API key on 22 November 2026, entering at not recorded. A model that did not answer has no page. How we measure.
Data checked · Last changed