Host DigitalOcean repriced gpt-oss-120b: cache read down 40%, input and output up 9%
Published 14 September 2026
Price move, per 1M tokens
- DigitalOcean · input↑ 9%$0.055 → $0.060
- DigitalOcean · output↑ 9%$0.385 → $0.420
- DigitalOcean · cache read↓ 40%$0.020 → $0.012
| host | rate | was | now | change |
|---|---|---|---|---|
| DigitalOcean | input | $0.055 | $0.060 | ↑ 9% |
| DigitalOcean | output | $0.385 | $0.420 | ↑ 9% |
| DigitalOcean | cache read | $0.020 | $0.012 | ↓ 40% |
input and output: ↑ 9%cache read: ↓ 40%
If you buy
If your workload leans on cached context, this host now suits you better, but re-check your input and output spend before the next top-up.
The model
gpt-oss-120bopen the model →
Open weights120.4B parameters131K context26 hosts
intelligence122nd of 168 writing138th of 168 coding126th of 168
gpt-oss-120b is a downloadable text model you can run yourself, because only a fraction of its parameters work on each token. Its licence allows commercial use, changes and redistribution, and its measured quality sits mid-field to lower-field on the preference boards.
Source
Read on openrouter.ai · we published this on 14 September 2026.
read the original at openrouter.ai ↗