Supported models
The DIAL Bedrock Adapter serves a fixed list of Amazon Bedrock models, addressed by their Bedrock model IDs in the adapter URL. This reference lists them with the features each supports. It is for platform engineers who choose which Bedrock models to offer in DIAL; the list follows the adapter's development branch, so check the release you run.
Chat models
Every model below serves POST /openai/deployments/<model id>/chat/completions.
| Vendor | Models | Model IDs | Input | Tools |
|---|---|---|---|---|
| Anthropic | Claude 5 Opus, Sonnet, and Fable | anthropic.claude-opus-5, anthropic.claude-sonnet-5, anthropic.claude-fable-5 | Text, images, documents | Yes |
| Anthropic | Claude 4.8, 4.7, and 4.6 Opus; 4.6 Sonnet | anthropic.claude-opus-4-8, anthropic.claude-opus-4-7, anthropic.claude-opus-4-6-v1, anthropic.claude-sonnet-4-6 | Text, images, documents | Yes |
| Anthropic | Claude 4.5 Sonnet and Haiku | anthropic.claude-sonnet-4-5-20250929-v1:0, anthropic.claude-haiku-4-5-20251001-v1:0, anthropic.claude-haiku-4-5 | Text, images, documents | Yes |
| Anthropic | Claude 4.1 Opus, 4 Opus, 4 Sonnet | anthropic.claude-opus-4-1-20250805-v1:0, anthropic.claude-opus-4-20250514-v1:0, anthropic.claude-sonnet-4-20250514-v1:0 | Text, images, documents | Yes |
| Anthropic | Claude 3.7 Sonnet, 3.5 Sonnet, 3.5 Sonnet 2.0 | anthropic.claude-3-7-sonnet-20250219-v1:0, anthropic.claude-3-5-sonnet-20240620-v1:0, anthropic.claude-3-5-sonnet-20241022-v2:0 | Text, images, documents | Yes |
| Anthropic | Claude 3.5 Haiku; Claude 3 Sonnet, Haiku, Opus | anthropic.claude-3-5-haiku-20241022-v1:0, anthropic.claude-3-sonnet-20240229-v1:0, anthropic.claude-3-haiku-20240307-v1:0, anthropic.claude-3-opus-20240229-v1:0 | Text and images, or text and documents | Yes |
| Amazon | Nova Pro, Nova Lite | amazon.nova-pro-v1:0, amazon.nova-lite-v1:0 | Text, images, documents | Yes |
| Amazon | Nova Micro | amazon.nova-micro-v1:0 | Text | No |
| Meta | Llama 4 Scout and Maverick 17B | meta.llama4-scout-17b-instruct-v1:0, meta.llama4-maverick-17b-instruct-v1:0 | Text, images | Yes |
| Meta | Llama 3.3 70B, 3.1 405B, 3.1 70B | meta.llama3-3-70b-instruct-v1:0, meta.llama3-1-405b-instruct-v1:0, meta.llama3-1-70b-instruct-v1:0 | Text | Yes |
| Meta | Llama 3.2 90B and 11B | meta.llama3-2-90b-instruct-v1:0, meta.llama3-2-11b-instruct-v1:0 | Text, images | 90B only |
| Meta | Llama 3.2 3B and 1B, 3.1 8B, Llama 3 70B and 8B | meta.llama3-2-3b-instruct-v1:0, meta.llama3-2-1b-instruct-v1:0, meta.llama3-1-8b-instruct-v1:0, meta.llama3-70b-instruct-v1:0, meta.llama3-8b-instruct-v1:0 | Text | No |
| DeepSeek | DeepSeek R1 | deepseek.r1-v1:0 | Text | No |
| MiniMax | MiniMax M2.5 | minimax.minimax-m2.5 | Text | Yes |
| AI21 Labs | Jamba 1.5 Large and Mini | ai21.jamba-1-5-large-v1:0, ai21.jamba-1-5-mini-v1:0 | Text | Yes |
| Cohere | Command R and R+ | cohere.command-r-v1:0, cohere.command-r-plus-v1:0 | Text, documents | Yes |
| Stability AI | Stable Diffusion 3.5 Large | stability.sd3-5-large-v1:0 | Text, images; returns an image | No |
| Stability AI | Stable Image Ultra and Core 1.0 | stability.stable-image-ultra-v1:1, stability.stable-image-core-v1:1 | Text; returns an image | No |
Every chat model supports /configuration. Claude uses the Anthropic SDK or the Converse API, Stability AI models the original Bedrock API, and the others the Converse API.
Token counting
Every chat model supports /truncate_prompt and the max_prompt_tokens parameter, and every model except the Stability AI ones supports /tokenize. Accuracy differs:
- Exact for Claude 5, Claude 4.8 and 4.7 Opus, and the Mantle Claude 4.5 Haiku (
anthropic.claude-haiku-4-5), when they use themantleclient. See Claude clients. - Approximate for every other model, because the vendors do not publish their tokenizers. The adapter counts each byte of the UTF-8 text as one token, which overestimates.
Cross-region inference
Models that support cross-region inference also take the region prefixes us., eu., and apac., for example us.anthropic.claude-3-5-sonnet-20241022-v2:0. Check that your AWS account supports cross-region inference for the model first.
Embedding models
These models serve POST /openai/deployments/<model id>/embeddings.
| Model | Model ID | Input |
|---|---|---|
| Titan Multimodal Embeddings G1 | amazon.titan-embed-image-v1 | Text, images |
| Titan Text Embeddings V2 | amazon.titan-embed-text-v2:0 | Text |
| Titan Embeddings G1 – Text v1.2 | amazon.titan-embed-text-v1 | Text |
| Cohere Embed English | cohere.embed-english-v3 | Text |
| Cohere Embed Multilingual | cohere.embed-multilingual-v3 | Text |
A model not on the list
When Bedrock releases a model the adapter does not know yet, serve it through a compatible model the adapter does know. See Compatibility mode.
Further reading
- DIAL Bedrock Adapter capabilities — configuration, guardrails, caching, and the Claude clients
- DIAL Bedrock Adapter configuration — AWS credentials and every adapter environment variable
- Supported providers — models across all DIAL adapters