Titan Models by Amazon Web Services (AWS) AI
About
The Amazon Titan family encompasses a suite of advanced large language models (LLMs) offered exclusively via Amazon Bedrock 145. These models are rooted in Amazon's extensive 25 years of AI and machine learning expertise 19, featuring high-performance text, multimodal, and image generation capabilities accessible through a fully managed API 19. The family includes models like Titan Text Premier, Express, and Lite, each catering to specific enterprise needs and optimizing cost and performance 145. Additionally, multimodal and image models such as Titan Multimodal Embeddings and Titan Image Generator are part of the suite 113. Responsible AI practices are integral, with models equipped to identify and filter harmful content 1.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
Use when provider availability and model metadata match the workload.
Use when the workload needs 32k context and structured outputs.
Use when the workload needs 8k context and structured outputs.
Use when the workload needs 4k context and structured outputs.
Use when provider availability and model metadata match the workload.
Use when the workload needs 8k context and structured outputs.
Use when the workload needs 4k context and structured outputs.
Use when provider availability and model metadata match the workload.
Use when the workload needs structured outputs.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Amazon Titan Image Generator G1 v2 | Use when provider availability and model metadata match the workload. | 2024-06 | — | Current |
| Titan Text Premier | Use when the workload needs 32k context and structured outputs. | 2024-05 | 32k contextstructured outputs | Current |
| Amazon Titan Text G1 - Express | Use when the workload needs 8k context and structured outputs. | 2024-05 | 8k contextstructured outputs | Current |
| Amazon Titan Text G1 - Lite | Use when the workload needs 4k context and structured outputs. | 2024-05 | 4k contextstructured outputs | Current |
| Amazon Titan Embeddings G1 - Text | Use when the workload needs 512 context. | 2024-04 | 512 context | Current |
| Amazon Titan Text Embeddings V2 | Use when provider availability and model metadata match the workload. | 2024-01 | — | Current |
| Titan Text Express | Use when the workload needs 8k context and structured outputs. | 2023-11 | 8k contextstructured outputs | Current |
| Titan Text Lite | Use when the workload needs 4k context and structured outputs. | 2023-11 | 4k contextstructured outputs | Current |
| Amazon Titan Multimodal Embeddings G1 | Use when provider availability and model metadata match the workload. | 2023-11 | — | Current |
| Amazon Titan Text Large | Use when the workload needs structured outputs. | 2023-11 | structured outputs | Current |
Release Timeline
5 release groupsSpecifications(10 models)
| Model | Released | Context | Structured Outputs |
|---|---|---|---|
| Amazon Titan Image Generator G1 v2 | 2024-06 | — | No |
| Titan Text Premier | 2024-05 | 32k | Yes |
| Amazon Titan Text G1 - Express | 2024-05 | 8k | Yes |
| Amazon Titan Text G1 - Lite | 2024-05 | 4k | Yes |
| Amazon Titan Embeddings G1 - Text | 2024-04 | 512 | No |
| Amazon Titan Text Embeddings V2 | 2024-01 | — | No |
| Titan Text Express | 2023-11 | 8k | Yes |
| Titan Text Lite | 2023-11 | 4k | Yes |
| Amazon Titan Multimodal Embeddings G1 | 2023-11 | — | No |
| Amazon Titan Text Large | 2023-11 | — | Yes |
Available From(2 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Amazon Titan Text Large | AWS Bedrock | $0.02 | $0.08 | Serverless |
| Amazon Titan Text Embeddings V2 | Vercel AI Gateway | $0.02 | — | Serverless |
| Amazon Titan Embeddings G1 - Text | AWS Bedrock | $0.1 | — | Serverless |
| Titan Text Lite | AWS Bedrock | $0.15 | $0.2 | Serverless |
| Titan Text Express | AWS Bedrock | $0.2 | $0.6 | Serverless |
| Amazon Titan Text G1 - Lite | AWS Bedrock | $0.4 | $0.8 | Serverless |
| Titan Text Premier | AWS Bedrock | $0.5 | $1.5 | Serverless |
| Amazon Titan Text G1 - Express | AWS Bedrock | $0.8 | $1.6 | Serverless |
Frequently Asked Questions
- What is Titan used for?
- Titan is used for structured outputs, coding, and agent workflows. The family description and listed model capabilities point to those workloads as the best fit.
- How does Titan compare to Amazon Nova?
- Titan by Amazon Web Services (AWS) AI is strongest where you need structured outputs, while Amazon Nova by Amazon Web Services (AWS) AI is the closest related family to check for multimodal. Titan has 10 listed variants and reaches up to 32k context, while Amazon Nova reaches up to 1m context, so compare the specs and pricing tables before choosing a production model.
- Which Titan model should I use?
- For the lowest listed input price, start with Amazon Titan Text Large through AWS Bedrock at $0.02/1M input tokens. For the most capable/latest local choice, evaluate Titan Text Premier with 32k context and structured outputs.
Related Model Families
Models(10)
Amazon Titan Image Generator G1 v2
Titan Text Premier
Amazon Titan Text G1 - Express
Amazon Titan Text G1 - Lite
Amazon Titan Embeddings G1 - Text
Amazon Titan Text Embeddings V2
Titan Text Express
Titan Text Lite
Amazon Titan Multimodal Embeddings G1
Amazon Titan Text Large