Google Gemini API
Google developer API for language, image, video, speech and music models; distinct from Vertex AI deployments.
Models on Google Gemini API 12
A curated selection, with links to other providers for the same model.
Explore imported models by type. Image and video categories refer to generation; image understanding is available under Vision.
Artificial Analysis Intelligence Index ↗ · Higher is better · Checked 2026-10-02
Rank among verified models in this catalog. Best published configuration per model; equal scores share a rank. Scores are rounded by the source. Unverified and non-language models appear last without a rank. Methodology ↗12 models
| Rank | Model | Intelligence | Best price | Input → output | Providers |
|---|---|---|---|---|---|
| #3 | Gemini 4 ArgonGoogle | 53high | $2 in / $10 outUSD / 1M tokensGoogle Gemini API Announced introductory · Lowest listed input price; output is from the same tariff. Announced launch pricing. Limited trusted-tester rollout; public API endpoint and launch date are not yet confirmed. Introductory period end date is unspecified. Post-introductory tariff is future pricing, not currently effective. Compare prices → | text, image, video → text | Google Gemini API |
| #17 | Gemini 3.8 FlashGoogle | 41high | $0.375 in / $1.875 outUSD / 1M tokensGoogle Gemini API Batch · Lowest listed input price; output is from the same tariff. Input: $0.375 through December 31, 2026. $0.75 starting January 1, 2027. Output: $1.875 through December 31, 2026. $3.75 starting January 1, 2027.. Cache, grounding and storage billed separately. Compare prices → | text, image, video, audio, PDF → text | Google Cloud Vertex AIGoogle Gemini APIOpenRouter |
| — | Gemini 3.8 LiveGoogle | —Not comparable | Price not collected | text, image, audio, video → audio, text (transcription) | Google Gemini API |
| — | Gemini Omni FlashGoogle | —Not comparable | Price not collected | text, image, video, audio → video, audio | Google Gemini API |
| — |  Lyria 3 ClipGoogle | —Not comparable | Price not collected | text, image → audio, text (lyrics) | Google Gemini API |
| — | Lyria 3.5Google | —Not comparable | Price not collected | text, image → audio, text (lyrics) | Google Gemini API |
| — | Lyria RealTimeGoogle | —Not comparable | Price not collected | text → audio | Google Gemini API |
| — |  Nano Banana 2Google | —Not comparable | $0.5 in / $3 outUSD / 1M tokensOpenRouter Router catalog · Lowest listed input price; output is from the same tariff. OpenRouter route catalog price, not a direct-provider price. Routing, modality and endpoint conditions can change the actual bill. Compare prices →From $0.067USD / imageReplicate target resolution=1K · Starting tariff; final cost depends on the selected settings. per output image Compare prices → | text, image, video, PDF → image, text | fal.aiGoogle Gemini APIOpenRouterReplicateRunway |
| — | Nano Banana 2 LiteGoogle | —Not comparable | Price not collected | text, image, video, PDF → image, text | Google Gemini API |
| — | Nano Banana ProGoogle | —Not comparable | Price not collected | text, image → image, text | Google Gemini API |
| — | Veo 3.1Google | —Not comparable | From $0.2USD / secondRunway veo3.1 (no audio) · Starting tariff; final cost depends on the selected settings. 20 credits per second Compare prices → | text, image → video, audio | Google Gemini APIRunwayTogether AI |
| — | Veo 3.1 LiteGoogle | —Not comparable | Price not collected | text, image → video, audio | Google Gemini API |
Catalog entries link to their specification and availability sources. This is a curated selection, not every model each provider offers. Best prices are the lowest collected USD tariffs, not a total-cost estimate. Token offers are selected by input price, with output from the same tariff; workload mix may change the cheapest provider. Listed variants may require batch processing, off-peak hours, or specific media settings. Price coverage varies by provider; collected tariffs include their source and observation time. Benchmark scores are a dated snapshot, not a live feed or a universal measure of quality. They apply to the shown evaluation configuration; listed prices may use different settings. Models without verified scores are not assumed to be weaker.