Imagen 4 Fast
Imagen 4 Fast is Google's throughput-optimized image generation model, the fast tier of Imagen 4, built for high-volume pipelines, real-time applications, and rapid creative iteration where generation speed and cost per image determine whether a workflow is viable.
import { experimental_generateImage as generateImage } from 'ai';
const result = await generateImage({ model: 'google/imagen-4.0-fast-generate-001', prompt: 'A red balloon on a wooden table.'});Playground
Try out Imagen 4 Fast by Google. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Providers
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
More models by Google
| Model |
|---|
| 1M | 2.1s | 299tps | $1.50/M | $9.00/M | Read:$0.15/M Write:— | $14.00/K + input costs | — |   | 05/19/2026 | ||||
| 1M | 0.8s | 255tps | $0.25/M | $1.50/M | Read:$0.03/M Write:— | $14.00/K + input costs | — | | 03/03/2026 | ||||
| 1M | 2.1s | 136tps | $2.00/M | $12.00/M | Read: $0.2/M Write: — | $14.00/K + input costs | — | | 02/19/2026 | ||||
| 1M | 0.7s | 185tps | $0.50/M | $3.00/M | Read: $0.05/M Write: — | $14.00/K + input costs | — | | 12/17/2025 | ||||
| 1M | 0.3s | 321tps | $0.10/M | $0.40/M | Read:$0.01/M Write:— | $35.00/K + input costs | — | | 06/17/2025 | ||||
| 1M | 0.4s | 189tps | $0.30/M | $2.50/M | Read:$0.03/M Write:— | $35.00/K + input costs | — | | 03/20/2025 |
About Imagen 4 Fast
Imagen 4 Fast is the speed-optimized tier in Google's Imagen 4 family. All three Imagen 4 variants (fast, standard, and ultra) launched simultaneously on AI Gateway as dedicated image-only models. The fast tier prioritizes generation speed and cost efficiency over maximum visual fidelity.
For interactive applications like product configurators, creative tools with live preview, and social media features where users expect near-real-time image generation, the fast tier's lower latency makes those features viable. For batch pipelines that process large catalogs of product images, generate thumbnails at scale, or apply visual transformations to a content library, the cost-per-image economics can determine whether a workflow is financially sustainable.
Because it's a pure image generation model, the integration surface is minimal. A text prompt goes in, image data comes out. There's no text output channel to parse, no modality configuration to manage, and no reasoning layer overhead. Teams building image pipelines can drop fast tier generates into existing workflows without architectural changes.
The fast tier delivers Imagen 4 quality. The fast-versus-standard distinction is about where on the speed-fidelity curve the model sits, not whether it produces usable output. For most high-volume use cases, the fast tier produces production-ready images.
What To Consider When Choosing a Provider
- Configuration: This is a text-to-image model only: use
experimental_generateImage(aliased asgenerateImage) from the AI SDK. It does not supportgenerateText, streaming chat, or any response modality beyond image output. - Zero Data Retention: AI Gateway supports Zero Data Retention for this model via direct gateway requests (BYOK is not included). To configure this, check the documentation.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
When to Use Imagen 4 Fast
Best For
- High-volume image pipelines: Per-image cost and generation throughput determine operational feasibility
- Real-time generation features: Interactive experiences require fast response times
- Concurrent batch jobs: Processing large prompt sets or product catalogs in parallel
- Iterative creative tools: Designers iterate through dozens of variations per session and turnaround speed affects workflow efficiency
- Cost-sensitive production: Quality requirements are met by the fast tier and the standard or ultra premium is unjustified
Consider Alternatives When
- Output quality is the bottleneck: Generation speed is not the constraint (consider
google/imagen-4.0-generate-001) - Maximum visual detail required: The use case demands the most detailed Imagen 4 output (consider
google/imagen-4.0-ultra-generate-001) - Text output alongside images: A multimodal model is needed (consider
google/gemini-3.1-flash-image-previeworgoogle/gemini-3-pro-image) - Video output required: Still images are not sufficient (consider the Veo model family)
Conclusion
Imagen 4 Fast is the high-throughput endpoint in the Imagen 4 family. For applications where generation volume, response latency, or cost per image are binding constraints, it delivers Imagen 4 quality within a speed and cost profile designed for production-scale operation.
Frequently Asked Questions
What is the realistic throughput with Imagen 4.0 Fast on AI Gateway?
Throughput depends on provider-specific rate limits and concurrent request capacity. AI Gateway supports parallel requests, and the fast tier's lower per-request latency allows more generations per unit of time compared to standard or ultra.
Can I generate multiple images concurrently in a batch job?
Yes. Fire concurrent
generateImagerequests against AI Gateway. The fast tier's lower latency means batch jobs complete faster, and AI Gateway handles retries and failover automatically for reliability under load.What is the cost difference between fast, standard, and ultra tiers?
Per-image costs appear in the Specs table above. The fast tier is priced below standard, which is priced below ultra. For large-volume workloads, the cost differential compounds significantly. Choose the fast tier if its quality meets your requirements.
Does Imagen 4.0 Fast accept image inputs for editing?
Imagen 4.0 Fast is a text-to-image generation model. For image editing or inpainting workflows, review the AI Gateway model catalog for models that explicitly support image-to-image operations.
Do I need a Google Cloud account to use this model?
No. AI Gateway manages all provider credentials. Authenticate using a Vercel API key or OIDC token.
Can I test the fast tier in the playground to compare it against standard or ultra before committing?
Yes. Visit https://ai-sdk.dev/playground/vertex:imagen-4.0-fast-generate-001 to generate images from any Imagen 4 tier side-by-side without writing integration code. This is the recommended way to evaluate whether the fast tier's output meets your quality bar before committing.
Is the fast tier appropriate for production deployments?
Yes. The fast tier is designed for production throughput, not just prototyping. AI Gateway adds observability, retries, and failover that further support reliable production operation.
What happens if I need higher quality for some images in an otherwise fast-tier pipeline?
AI Gateway's unified API makes it straightforward to route specific requests to a different model tier. Use
google/imagen-4.0-generate-001orgoogle/imagen-4.0-ultra-generate-001for requests that require higher fidelity, with no infrastructure change required.