Content Studio AI models
Different AI models will generate varying content, so be sure to try them all to see which one generates the kind of content you're looking for. Feel free to switch between them depending on what works for your brand.
AI models are constantly evolving, which means within Content Studio models can move between different lifecycle stages: Labs, Supported, Legacy and Retired. Read more about our model lifecycle stages.
Available models
Last updated: March 2025
The AI models that you can select to use with Content Studio are:
| Model | Provider | Capabilities | Description | Lifecycle stage | Planned retirement date | *Available for automated services |
|---|---|---|---|---|---|---|
| GPT 4o | Open AI | Reads images | A versatile, high-intelligence flagship model. It accepts both text and image inputs, and produces text outputs. | Supported | Yes | |
| Claude 3.7 Sonnet | Anthropic | Reads images | Anthropic’s most intelligent model provides top-tier results in reasoning, coding, multilingual tasks, long-context handling, honesty, and image processing. Produces more human-like responses. | Supported | Yes | |
| Claude 3.5 Sonnet | Anthropic | Reads images | Anthropic's previous most intelligent model - high level of intelligence and capability. | Supported | Yes | |
| Claude 3 Opus | Anthropic | Reads images | Powerful model for complex tasks - top-level intelligence, fluency, and understanding. | Legacy | July 1 2025 | No |
| Meta Llama 3.1 | Meta | Text only | Meta's flagship, multilingual text model. | Legacy | July 1 2025 | No |
*Models are available for automated services only when their Lifecycle stage is "Supported".
Retired models
View the list of retired AI models
| AI model | Description | Retired date |
|---|---|---|
| GPT-4 | OpenAI's large, high quality model, knowledge up to September 2021 | March 4 2025 |
| GPT-4 Turbo | OpenAI's large, fast and cost-efficient model, knowledge up to December 2023 | March 4 2025 |
| GPT-3.5 | OpenAI's high-speed model, knowledge up to September 2021 | August 29 2024 |
Model lifecycle stages
AI models used in Amplience apps and services will be in one of the following stages:
- Labs
- Supported
- Legacy
- Retired
Labs
Models labeled as Labs are designed for experimentation and provide you with early access to innovations that you can evaluate for use in your workflow.
Supported
Models classified as Supported, are available and suitable for use in production environments, and can be used to configure custom templates, integrations, automations. Models in the Supported stage will remain available until the Amplience retirement date - which is based on deprecation dates from the respective model providers.
Legacy
Once an Amplience retirement date is set, the model moves into a Legacy stage. Legacy models can still be used for generating content where they are already in place, but need to be substituted in integrations and other services before the retirement date. You should plan to move to a different model while a model is in the legacy stage.
Retired
Models labeled Retired are no longer available for use.
Note: Amplience does not guarantee service for integrations when in-use AI models become retired.