AI Rate Carddavinci-002 vs GPT-3.5-turbo: API price and specs
GPT-3.5-turbo is 88% cheaper than davinci-002 on input price.
GPT-3.5-turbo is 62% cheaper than davinci-002 on output price.
GPT-3.5-turbo is 83% cheaper than davinci-002 on a 1M-in / 200k-out workload.
| Field | davinci-002 | GPT-3.5-turbo |
|---|---|---|
| Lowest input, per 1M | $2.00 +700% | $0.25 |
| Median input, per 1M | $2.00 +300% | $0.50 |
| Lowest output, per 1M | $2.00 +166.7% | $0.75 |
| Median output, per 1M | $2.00 +33.3% | $1.50 |
| Lowest unit price | — | — |
| Cost of 1M in + 200k out | $2.40 +500% | $0.40 |
| Context | — | 16K |
| Max output | 16K | 4K +75% |
| Providers | 2 +83.3% | 12 |
| Regions | — | US |
| Accepts | text | image, pdf, text |
| Returns | text | text |
| Type | Text models | Text models |
| Free option | no | no |
| Cheapest provider | aihubmix | openrouter |
Your own workload: how many tokens the model gets (input) and gives back (output) - for example, in a month. The "Cost" row of the table above is recalculated for these numbers, for each model.
Computed from current offers each time the page opens; the best value of a row is bold, the percent is how far behind the best.
Providers selling both
Frequently asked questions
Which is cheaper, davinci-002 or GPT-3.5-turbo?
GPT-3.5-turbo is 88% cheaper than davinci-002 on input price. GPT-3.5-turbo is 62% cheaper than davinci-002 on output price. GPT-3.5-turbo is 83% cheaper than davinci-002 on a 1M-in / 200k-out workload.
Where can I use both davinci-002 and GPT-3.5-turbo?
aihubmix, openai