openai/gpt-3.5-turbo vs openai/gpt-oss-20b
For 100,000 calls of 2,000 in and 500 out, openai/gpt-oss-20b is cheaper: $8.10 a month at OpenRouter, against $175.00 for openai/gpt-3.5-turbo at OpenRouter — 21.6× the cost.
Cheapest on each
openai/gpt-3.5-turbo
$175.00/mo
cheapest, at OpenRouter — $0.5 in, $1.5 out per million
openai/gpt-oss-20b
$8.10/mo
cheapest, at OpenRouter — $0.018 in, $0.09 out per million
Aligned facts
Same rows on both sides, including the ones we cannot fill.
| Attribute | openai/gpt-3.5-turbo | openai/gpt-oss-20b |
|---|---|---|
| Context window | 16K | 131K |
| Accepts | Aa | Aa |
| Sellers | 1 | 1 |
| Cheapest input rate | $0.5 / M | $0.018 / M |
| Cached input rate | ○ no answer | $0.009 / M |
No quality or benchmark row appears here. We hold no throughput or accuracy measurement for either model, and a cost-per-quality figure built from a lab’s own claim would look authoritative and mean nothing.
Cost for this workload — both models, every seller
100,000 calls a month · 2,000 input and 500 output tokens each. Sellers of both models lead, cheapest combined cost first.
| Seller | openai/gpt-3.5-turbo /mo | openai/gpt-oss-20b /mo |
|---|---|---|
| OpenRouter | $175.002026-10-24 | $8.102026-10-27 |
Cheapest tier per seller, this workload only. In and output rates, and a cached prefix, are broken out in the cost calculator.