Gemini Flash
FamilyGoogle · Gemini
Google's fast Gemini line, led by Gemini 3.5 Flash plus Live, TTS, Translate, and efficient variants.
Overview
This is a model family overview. For version-specific details, see the individual model entries linked below.
Gemini Flash is Google’s speed-and-cost tier, designed for tasks where throughput, latency, and price matter alongside strong multimodal reasoning. The family now spans stable production models, preview fast-model experiments, Live/TTS/Translate variants, and adjacent browser-control or agent surfaces. The center of gravity moved in May 2026 with Gemini 3.5 Flash, Google’s stable fast frontier model for coding, multimodal understanding, and long-horizon agent workflows.
Current Latest
Gemini 3.5 Flash is the current stable fast/agentic route in the Gemini API. Google’s current catalog also keeps Gemini 3 Flash as a preview entry with Gemini 3.5 Flash as the recommended replacement, Gemini 3.1 Flash-Lite as the stable efficient route, Gemini 3.1 Flash Live/TTS as adjacent preview audio surfaces, and Gemini 3.5 Live Translate as a narrower realtime translation route. Older Gemini 2.5 Flash entries remain useful compatibility and cost baselines, but Google’s deprecation page now lists October 16, 2026 shutdown dates for the stable 2.5 Flash and Flash-Lite endpoints.
Strengths
- Very fast inference for latency-sensitive applications
- Stable Gemini 3.5 Flash route for agentic coding and long-horizon workflows
- Competitive pricing relative to Pro tiers
- Full multimodal support across text, image, video, audio, and PDFs
- 1M-token context windows on stable Flash and Flash-Lite
- Stable Flash-Lite variant for the most cost-sensitive high-volume workloads
- Preview-tier Gemini 3 Flash and Live/TTS/Translate variants for teams tracking newer fast-model direction
When to Choose Gemini Flash
- High-volume processing where cost per request matters
- Real-time applications requiring low latency
- Bulk document analysis and extraction pipelines
- Development prototyping before escalating to Pro or managed-agent routes
- Applications where multimodal support is needed at scale
- Teams that want a stable 3.5 Flash production lane while evaluating preview and media-specific variants
Access
- Google AI Studio
- Vertex AI and Gemini Enterprise Agent Platform deployment paths
- Google Gemini consumer products
- Third-party integrations via API
Model Versions
Gemini 3.5 Live Translate
previewGoogle's realtime speech-to-speech translation model for the Gemini Live API.
Gemini 3.5 Flash
currentGoogle's stable Gemini 3.5 Flash model for fast frontier multimodal, coding, and long-horizon agent workflows.
Gemini 3.1 Flash-Lite
currentGoogle's stable low-cost Gemini 3.1 Flash-Lite tier for high-throughput multimodal assistant and automation workloads.
Gemini 3.1 Flash TTS Preview
previewGoogle's expressive Gemini 3.1 Flash TTS preview, an API-first text-to-speech model spanning 70+ languages.
Gemini 3.1 Flash Live Preview
previewGoogle's low-latency Gemini 3.1 live model for realtime audio-to-audio and multimodal dialogue.
Gemini 3 Flash
previewGoogle's older Gemini 3 preview Flash route, now superseded by stable Gemini 3.5 Flash for most new fast-model work.
Gemini 2.5 Flash Live Preview
previewGoogle's stable 2.5-era native-audio Live API model for realtime multimodal voice agents.
Gemini 2.5 Flash-Lite
currentStable budget Gemini 2.5 tier for large-scale assistant and automation workloads.
Gemini 2.5 Flash
currentStable Gemini 2.5 Flash route balancing multimodal capability, latency, and production cost.