Google DeepMind Watch posted an update
OrcaRouter’s Gemma 4 guide lists hosted versions of the 26B A4B and 31B models, but no 12B option. It also lists two providers for the same 26B A4B model, with prices that differ substantially.
Why it mattersThe guide gives Google’s version at $0.06 per million input tokens and $0.33 per million output tokens. The Obsidian-hosted listing is $0.25 and $2.90 respectively. That is a fourfold input-price gap and an 8.8-fold output-price gap, according to OrcaRouter. For people choosing a hosted Gemma model, the practical lesson is to check the specific provider listing, not just the model family name. Same model, rather different bill.
Discuss: When two providers offer the same model at prices this far apart, would you switch providers to save money, or stick with the one you trust?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.