Imagen Models by Google DeepMind
Google DeepMindProprietary
9 models2024
About
Google's Imagen family of image generation models, capable of producing high-quality, photorealistic images from text prompts. Includes Imagen 3.0 and 4.0 variants for different speed/quality trade-offs.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
9 in view
Imagen Product RecontextCurrent
Use when the workload needs image and multimodal inputs.
2024-10imagemultimodal inputs
Virtual Try-OnCurrent
Use when the workload needs image and multimodal inputs.
2024-09imagemultimodal inputs
Use when the workload needs image and multimodal inputs.
2024-08imagemultimodal inputs
Imagen 3Current
Use when the workload needs image and multimodal inputs.
2024-08imagemultimodal inputs
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Imagen 4 Ultra | Use when the workload needs image. | 2024-11 | image | Current |
| Imagen 4 | Use when the workload needs image. | 2024-11 | image | Current |
| Imagen 4 Fast | Use when the workload needs image. | 2024-11 | image | Current |
| Imagen Product Recontext | Use when the workload needs image and multimodal inputs. | 2024-10 | imagemultimodal inputs | Current |
| Virtual Try-On | Use when the workload needs image and multimodal inputs. | 2024-09 | imagemultimodal inputs | Current |
| Imagen 3 | Use when the workload needs image. | 2024-08 | image | Current |
| Imagen 3 Fast | Use when the workload needs image. | 2024-08 | image | Current |
| Imagen 3 for Editing and Customization | Use when the workload needs image and multimodal inputs. | 2024-08 | imagemultimodal inputs | Current |
| Imagen 3 | Use when the workload needs image and multimodal inputs. | 2024-08 | imagemultimodal inputs | Current |
Release Timeline
4 release groups2024-11
3 current
2024-10
1 current
Imagen Product Recontext
Currentimagemultimodal inputs
2024-09
1 current
Virtual Try-On
Currentimagemultimodal inputs
2024-08
4 current
Imagen 3
Currentimage
Imagen 3
Currentimagemultimodal inputs
Imagen 3 Fast
Currentimage
Imagen 3 for Editing and Customization
Currentimagemultimodal inputs
Specifications(9 models)
| Model | Released | Vision | Multimodal |
|---|---|---|---|
| Imagen 4 Ultra | 2024-11 | No | No |
| Imagen 4 | 2024-11 | No | No |
| Imagen 4 Fast | 2024-11 | No | No |
| Imagen Product Recontext | 2024-10 | Yes | Yes |
| Virtual Try-On | 2024-09 | Yes | Yes |
| Imagen 3 | 2024-08 | No | No |
| Imagen 3 Fast | 2024-08 | No | No |
| Imagen 3 for Editing and Customization | 2024-08 | Yes | Yes |
| Imagen 3 | 2024-08 | Yes | Yes |
Available From(2 providers)
Frequently Asked Questions
- What is Imagen used for?
- Imagen is used for image and vision and multimodal work. The family description and listed model capabilities point to those workloads as the best fit.
- How does Imagen compare to Gemma 4?
- Imagen by Google DeepMind is strongest where you need image, while Gemma 4 by Google DeepMind is the closest related family to check for vision and multimodal work. Imagen has 9 listed variants, while Gemma 4 reaches up to 256K context, so compare the specs and pricing tables before choosing a production model.
- Which Imagen model should I use?
- If price is the main constraint, use the pricing table first because Imagen does not have complete provider pricing in the local data. For the most capable/latest local choice, evaluate Imagen Product Recontext with multimodal inputs.
Related Model Families
Models(9)
Imagen 4 Ultra
2024-112 providers
Imagen 4
2024-112 providers
Imagen 4 Fast
2024-112 providers
Imagen Product Recontext
2024-101 provider
Multimodal
Virtual Try-On
2024-091 provider
Multimodal
Imagen 3
2024-081 provider
Imagen 3 Fast
2024-081 provider
Imagen 3 for Editing and Customization
2024-081 provider
Multimodal
Imagen 3
2024-081 provider
Multimodal