Gemini 2.5 Flash

Gemini 2.5 Flash is a fast and cost-effective model that balances performance with a wide range of capabilities. It is the first Flash model to feature thinking capabilities, which lets you see the model's thinking process as it generates a response.

The Gemini 2.5 Flash model is available in two versions:

Model Version Description Use Case
Standard Gemini 2.5 Flash The standard version of the model, which offers a balance of price, performance, and a wide range of capabilities, including thinking features. Use for general-purpose tasks that require a fast, cost-effective model with strong reasoning abilities.
Gemini 2.5 Flash Image Preview Our standard model upgraded for rapid creative workflows with image generation and conversational, multi-turn editing capabilities.
  • Generate interleaved text and images
  • Multi-turn image editing in natural language
  • Locale-aware image generation"
Gemini 2.5 Flash with Live API native audio (Preview) A preview version with advanced, native audio functionality that includes enhanced voice quality, proactive audio responses, and affective dialog capabilities. Use for applications that require sophisticated, natural, and emotionally aware voice interactions, such as advanced voice assistants or customer service bots.

For detailed technical information about performance benchmarks, training datasets, and our approach to safety, see the technical report and the model card for Gemini 2.5 Flash.

2.5 Flash

Try in Vertex AI View in Model Garden (Preview) Deploy example app

Note: To use the "Deploy example app" feature, you need a Google Cloud project with billing and Vertex AI API enabled.
Model ID gemini-2.5-flash
Supported inputs & outputs
  • Inputs:
    Text, Code, Images, Audio, Video
  • Outputs:
    Text
Token limits
  • Maximum input tokens: 1,048,576
  • Maximum output tokens: 65,535 (default)
Capabilities
Usage types
Input size limit 500 MB
Technical specifications
Images
  • Maximum images per prompt: 3,000
  • Maximum image size: 7 MB
  • Supported MIME types:
    image/png, image/jpeg, image/webp
Documents
  • Maximum number of files per prompt: 3,000
  • Maximum number of pages per file: 1,000
  • Maximum file size per file for the API or Cloud Storage imports: 50 MB
  • Maximum file size per file for direct uploads through the console: 7 MB
  • Supported MIME types:
    application/pdf, text/plain
Video
  • Maximum video length (with audio): Approximately 45 minutes
  • Maximum video length (without audio): Approximately 1 hour
  • Maximum number of videos per prompt: 10
  • Supported MIME types:
    video/x-flv, video/quicktime, video/mpeg, video/mpegs, video/mpg, video/mp4, video/webm, video/wmv, video/3gpp
Audio
  • Maximum audio length per prompt: Appropximately 8.4 hours, or up to 1 million tokens
  • Maximum number of audio files per prompt: 1
  • Speech understanding for: Audio summarization, transcription, and translation
  • Supported MIME types:
    audio/x-aac, audio/flac, audio/mp3, audio/m4a, audio/mpeg, audio/mpga, audio/mp4, audio/opus, audio/pcm, audio/wav, audio/webm
Parameter defaults
  • Temperature: 0.0-2.0 (default 1.0)
  • topP: 0.0-1.0 (default 0.95)
  • topK: 64 (fixed)
  • candidateCount: 1–8 (default 1)
Supported regions

Model availability

(Includes dynamic shared quota & Provisioned Throughput)

  • Global
    • global
  • United States
    • us-central1
    • us-east1
    • us-east4
    • us-east5
    • us-south1
    • us-west1
    • us-west4
  • Europe
    • europe-central2
    • europe-north1
    • europe-southwest1
    • europe-west1
    • europe-west4
    • europe-west8
    • europe-west9+

ML processing

  • United States
    • Multi-region
  • Canada
    • northamerica-northeast1
  • Europe
    • Multi-region
    • europe-west1*
  • Asia Pacific
    • asia-northeast1*
    • asia-northeast3*
    • asia-south1*
    • asia-southeast1
    • australia-southeast1*
See Data residency for more information.
Knowledge cutoff date January 2025
Versions
  • gemini-2.5-flash
    • Launch stage: GA
    • Release date: June 17, 2025
    • Discontinuation date: June 17, 2026
  • gemini-live-2.5-flash
    • Launch stage: Private GA
    • Release date: June 17, 2025
  • gemini-2.5-flash-preview-05-20
    • Launch stage: Public preview
    • Release date: May 20, 2025
    • Discontinuation date: July 15, 2025
  • gemini-2.5-flash-preview-04-17
    • Launch stage: Public preview
    • Release date: April 17, 2025
    • Discontinuation date: July 15, 2025
Security controls
See Security controls for more information.
Pricing See Pricing.
+ Supervised fine tuning not supported
* Available for 128K context window only

Image

Try in Vertex AI (Preview) Deploy example app

Note: To use the "Deploy example app" feature, you need a Google Cloud project with billing and Vertex AI API enabled.
Model ID gemini-2.5-flash-image-preview
Supported inputs & outputs
  • Inputs:
    Text, Images
  • Outputs:
    Text and image
Token limits
  • Maximum input tokens: 32,768
  • Maximum output tokens: 32,768
Capabilities
Usage types
Input size limit 500 MB
Technical specifications
Images
  • Maximum images per prompt: 3
  • Maximum image size: 7 MB
  • Maximum number of output images per prompt: 10
  • Supported MIME types:
    image/png, image/jpeg, image/webp
Documents
  • Maximum number of files per prompt: 3
  • Maximum number of pages per file: 3
  • Maximum file size per file: 50 MB
  • Supported MIME types:
    application/pdf, text/plain
Parameter defaults
  • Temperature: 0.0-2.0 (default 1.0)
  • topP: 0.0-1.0 (default 0.95)
  • topK: 64 (fixed)
  • candidateCount: 1–8 (default 1)
Supported regions

Model availability

  • Global
    • global
See Data residency for more information.
Knowledge cutoff date June 2025
Versions
  • gemini-2.5-flash-image-preview
    • Launch stage: Public preview
    • Release date: August 26, 2025
Security controls
See Security controls for more information.
Pricing See Pricing.

Live API native audio

The Gemini 2.5 Flash model with Live API native audio includes advanced audio functionality for Live API. In addition to the standard Live API features, this preview model includes the following:

  • Enhanced voice quality and adaptability: Live API native audio provides richer, more natural voice interactions with 30 HD voices in 24 languages.
  • Proactive Audio: When you enable Proactive Audio, the model responds only when relevant. It proactively generates text transcripts and audio responses only for queries directed at the device and ignores other queries.
  • Affective Dialog: Models using Live API native audio can understand and respond appropriately to a user's emotional expressions for more nuanced conversations.

For more information about Live API, see the standalone Live API documentation.

Try in Vertex AI

Model ID gemini-live-2.5-flash-preview-native-audio
Supported inputs & outputs
  • Inputs:
    Audio, Video
  • Outputs:
    Audio
Token limits
  • Maximum input tokens: 1,048,576
  • Maximum output tokens: 128K (default)
Capabilities
Usage types
Input size limit 500 MB
Technical specifications
Video
  • Maximum screenshare length: Approximately 10 minutes
  • Supported MIME types:
    video/x-flv, video/quicktime, video/mpeg, video/mpegs, video/mpg, video/mp4, video/webm, video/wmv, video/3gpp
Audio
  • Maximum conversation length: Approximately 10 minutes
  • Speech understanding for: Audio summarization, transcription, and translation
  • Supported MIME types:
    audio/x-aac, audio/flac, audio/mp3, audio/m4a, audio/mpeg, audio/mpga, audio/mp4, audio/opus, audio/pcm, audio/wav, audio/webm
Parameter defaults
  • Temperature: 0.0-2.0 (default 1.0)
  • topP: 0.0-1.0 (default 0.95)
  • topK: 64 (fixed)
  • candidateCount: 1–8 (default 1)
Supported regions

Model availability

  • United States
    • us-central1
See Data residency for more information.
Knowledge cutoff date January 2025
Versions
  • gemini-live-2.5-flash-preview-native-audio
    • Launch stage: Public preview
    • Release date: June 17, 2025
Security controls
See Security controls for more information.
Pricing See Pricing.