Philipp Schmid (Google DeepMind): gemini-nano-banana-2.1 at "$0.034 per image vs $0.134"

Posted 16:07 UTC on 6 Oct by Philipp Schmid, who works on agents and the Gemini API at Google DeepMind (his X bio). He gives the API id gemini-nano-banana-2.1 and says the model is "Outperforming previous Pro model at 4x cheaper price (now $0.034 per image vs $0.134)". Checked against the Gemini API prices recorded in our model files: $0.034 is the $0.0336 price of a 1K image, rounded, and $0.134 is the 1K/2K price of Nano Banana Pro (gemini-3-pro-image). So "4x cheaper" compares with Pro; against Nano Banana 2 ($0.067 per 1K image) the cut is half, as the timeline entry says. He also lists "up to 5 characters consistency + 14 references images", while the Gemini API model page (read 7 Oct) says 14 reference images with "character consistency for up to 4 characters and object fidelity for up to 10 objects". The mismatch is unresolved. Other claims: better instruction following and in-image text, grounding with Google Image Search; available in AI Studio, the Gemini API, Google Cloud, the Gemini app and Flow. Reach at 11:31 UTC on 7 Oct: ~30k views, 495 likes.

Nano Banana 2.1 (gemini-nano-banana-2.1) is now available. Outperforming previous Pro model at 4x cheaper price (now $0.034 per image vs $0.134).

🕒 Improved instruction following 🔤 Better in-image text rendering 🔍 Grounded with Google Image Search 🎭 up to 5 characters consistency + 14 references images

Available in @GoogleAIStudio, the Gemini API, @googlecloud, the @GeminiApp, and @FlowbyGoogle: https://aistudio.google.com/docs/image-generation

Media: https://pbs.twimg.com/media/HT9arOaXMAADozS.jpg?name=orig https://pbs.twimg.com/media/HT9arObWIAA4bpu.jpg?name=orig https://pbs.twimg.com/media/HT9arORXkAAucLC.jpg?name=orig https://pbs.twimg.com/media/HT9arOSXgAAL_CO.jpg?name=orig

views 29590 · likes 495 · reposts 35 · replies 34 · quotes 7 (at fetch time)

Archived 2026-10-07 via manual (api.fxtwitter.com, fetched 11:31 UTC).