Calling Gemini models for less
Gemini Flash is already one of the cheapest capable models on the market - the savings come from where the tokens are served from, not from swapping to a weaker model. We resell Google models at 30% below list, with the same request shape you already use.
Why Flash is usually the right default
- Long context at a low input price - good for pasting whole documents
- Fast enough for interactive UI without a spinner
- Strong multilingual output, which matters if your users are not English-first
Google list vs what we charge
| Model | Official in / out | Tapix AI in / out |
|---|---|---|
gemini-3-flash | $0.50 / $3.00 | $0.35 / $2.10 |
A note on honesty: some models in our catalogue are cheaper bought directly from the provider. Those are not pushed on this site - we only market models where we are genuinely below list.
Switching
base_url = https://tapixai.com/v1 and a Tapix AI key. Model IDs follow our catalogue, e.g. gemini-3-flash. Your prompts and parameters do not change.
Live prices from our catalogue
Read from our live catalogue on every page load, so these numbers cannot go stale. USD per 1M tokens.
| Model | Input / 1M | Output / 1M |
|---|---|---|
| — | — | — |