Llama 4 Scout → GPT-5 nano: cost per request on six tasks
GPT-5 nano costs between 47% less and 11% more than Llama 4 Scout per request on six common tasks. It costs no more per result that works if its results work first time at least 54% as often as Llama 4 Scout's on search re-ranking, and 11% more often on translation.
List prices per million tokens
| Model | Input | Cached input | Output |
|---|---|---|---|
| Llama 4 Scout | $0.10 | n/a | $0.30 |
| GPT-5 nano | $0.05 | $0.005 | $0.40 |
Prices checked 2026-11-12.
Cost per 1,000 requests
| Task | Llama 4 Scout | GPT-5 nano | Change | Cheaper per result that works if it works first time at least |
|---|---|---|---|---|
| Support bot answerInstructions, three help-centre passages and the customer's question in; a short answer out. | $0.39 | $0.27 | 31% less | 70% as often as Llama 4 Scout |
| Invoice field extractionInstructions and one invoice's text in; about 15 fields as JSON out. | $0.33 | $0.23 | 31% less | 70% as often as Llama 4 Scout |
| Lead scoringScoring rules and one lead's form answers in; a score and a one-line reason out. | $0.15 | $0.10 | 33% less | 67% as often as Llama 4 Scout |
| Email reply draftInstructions and the email thread so far in; a reply of about 300 words out. | $0.27 | $0.23 | 13% less | 88% as often as Llama 4 Scout |
| TranslationInstructions and about 500 words of text in; the same text in another language out. | $0.33 | $0.36 | 11% more | 11% more often than Llama 4 Scout |
| Search re-rankingA search query and 20 result snippets in; the ids of the best five, in order, out. | $0.41 | $0.22 | 47% less | 54% as often as Llama 4 Scout |
Token counts are typical sizes, not measurements.
Test it on your own prompt
The model switch test runs your prompt on both models side by side with your own API key, in your browser.