stepfun/step-3.5-flash vs stepfun/step-5-preview
For 100,000 calls of 2,000 in and 500 out, stepfun/step-3.5-flash is cheaper: $35.00 a month at OpenRouter, against $335.00 for stepfun/step-5-preview at Novita AI — 9.6× the cost.
Cheapest on each
stepfun/step-3.5-flash
$35.00/mo
cheapest, at OpenRouter — $0.1 in, $0.3 out per million
stepfun/step-5-preview
$335.00/mo
cheapest, at Novita AI — $1 in, $2.7 out per million
Aligned facts
Same rows on both sides, including the ones we cannot fill.
| Attribute | stepfun/step-3.5-flash | stepfun/step-5-preview |
|---|---|---|
| Context window | 262K | 1M |
| Accepts | Aa | Aa · IMG · VID |
| Sellers | 1 | 2 |
| Cheapest input rate | $0.1 / M | $1 / M |
| Cached input rate | ○ no answer | $0.1 / 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 | stepfun/step-3.5-flash /mo | stepfun/step-5-preview /mo |
|---|---|---|
| OpenRouter | $35.002026-10-27 | $335.002026-10-27 |
| Novita AI | ○ no answer | $335.002026-10-22 |
Cheapest tier per seller, this workload only. In and output rates, and a cached prefix, are broken out in the cost calculator.