Publié le 2026-08-16 · 赵予安
Réponse directe
Google Gemini 3 Flash Preview is listed at 0.5 / 3 per million tokens, with a 1,048,576 token context window. First-party cache scores are unpublished. Ce guide s’adresse aux équipes produit et plateforme qui comparent la qualité des modèles, le coût, la politique de routage et le risque de déploiement.
What shipped
Google's Gemini 3 Flash Preview is now listed in the NextModel gateway catalog as google/gemini-3-flash-preview. The public slug is google--gemini-3-flash-preview. The provider is Google. The display name is "Google: Gemini 3 Flash Preview".
Price and context
The listed input price is $0.5 per 1M tokens. The listed output price is $3 per 1M tokens. The model has a context window of 1,048,576 tokens. The catalog places latency in the 1000 to 3000ms band. For comparison, Gemini 2.5 Flash is listed at $0.30 per 1M input tokens and $2.50 per 1M output tokens. Gemini 3.5 Flash Lite has the same listed rates. Gemini 3.1 Flash Lite is listed at $0.25 per 1M input tokens and $1.50 per 1M output tokens.
Who this is for
The catalog lists vision, long-context, and agent work as the use cases. The model has capabilities for streaming, tool calling, JSON, vision, and long-context tasks. Developers building agent, vision, or long-context applications are the intended audience.
What is still unverified
The release date is unpublished in this catalog snapshot. Independent third-party quality scores are not in these notes. No first-party CacheSafety row exists for this exact model. Safe Hit, Bad Hit, and trap rates are not available.