feat: add BedrockEmbeddingModel for Nova, Cohere and Titan endpoints

docs/api/embeddings.md

@@ -14,6 +14,8 @@
 
 ::: pydantic_ai.embeddings.google
 
+::: pydantic_ai.embeddings.bedrock
+
 ::: pydantic_ai.embeddings.sentence_transformers
 
 ::: pydantic_ai.embeddings.test

docs/embeddings.md

@@ -342,6 +342,156 @@ embedder = Embedder(
 )
 ```
 
+### Bedrock
+
+[`BedrockEmbeddingModel`][pydantic_ai.embeddings.bedrock.BedrockEmbeddingModel] provides access to embedding models through AWS Bedrock, including Amazon Titan, Cohere, and Amazon Nova models.
+
+#### Install
+
+To use Bedrock embedding models, you need to either install `pydantic-ai`, or install `pydantic-ai-slim` with the `bedrock` optional group:
+
+```bash
+pip/uv-add "pydantic-ai-slim[bedrock]"
+```
+
+#### Configuration
+
+Authentication with AWS Bedrock uses standard AWS credentials. See the [Bedrock provider documentation](models/bedrock.md#environment-variables) for details on configuring credentials via environment variables, AWS credentials file, or IAM roles.
+
+Ensure your AWS account has access to the Bedrock embedding models you want to use. See [AWS Bedrock model access](https://docs.aws.amazon.com/bedrock/latest/userguide/model-access.html) for details.
+
+#### Basic Usage
+
+```python {title="bedrock_embeddings.py" test="skip"}
+from pydantic_ai import Embedder
+
+# Using Amazon Titan
+embedder = Embedder('bedrock:amazon.titan-embed-text-v2:0')
+
+
+async def main():
+    result = await embedder.embed_query('Hello world')
+    print(len(result.embeddings[0]))
+    #> 1024
+```
+
+_(This example requires AWS credentials configured)_
+
+#### Supported Models
+
+Bedrock supports three families of embedding models via [`BedrockEmbeddingSettings`][pydantic_ai.embeddings.bedrock.BedrockEmbeddingSettings]. See the [AWS Bedrock documentation](https://docs.aws.amazon.com/bedrock/latest/userguide/models-supported.html) for the full list of available models.
+
+##### Amazon Titan
+
+- `amazon.titan-embed-text-v1` — 1536 dimensions, 8K tokens
+- `amazon.titan-embed-text-v2:0` — 256/384/1024 dimensions (configurable), 8K tokens
+
+**Titan-specific settings:**
+
+- `bedrock_titan_normalize` — Normalize embedding vectors for direct similarity calculations.
+
+```python {title="bedrock_titan.py"}
+from pydantic_ai import Embedder
+from pydantic_ai.embeddings.bedrock import BedrockEmbeddingSettings
+
+embedder = Embedder(
+    'bedrock:amazon.titan-embed-text-v2:0',
+    settings=BedrockEmbeddingSettings(
+        bedrock_titan_normalize=True,
+    ),
+)
+```
+
+##### Cohere Embed
+
+- `cohere.embed-english-v3` — English-only, 512 tokens
+- `cohere.embed-multilingual-v3` — Multilingual, 512 tokens
+- `cohere.embed-v4:0` — Latest Cohere model, 128K tokens
+
+**Cohere-specific settings:**
+
+- `bedrock_cohere_max_tokens` — Maximum number of tokens to embed.
+- `bedrock_cohere_input_type` — Input type: `'search_query'`, `'search_document'`, `'classification'`, or `'clustering'`. Defaults to `'search_query'` for `embed_query()` and `'search_document'` for `embed_documents()`.
+- `bedrock_cohere_truncate` — Truncation strategy: `'NONE'` (default, raises error if input exceeds max tokens), `'START'`, or `'END'`. Overrides the base `truncate` setting if provided.
+
+```python {title="bedrock_cohere.py"}
+from pydantic_ai import Embedder
+from pydantic_ai.embeddings.bedrock import BedrockEmbeddingSettings
+
+embedder = Embedder(
+    'bedrock:cohere.embed-english-v3',
+    settings=BedrockEmbeddingSettings(
+        bedrock_cohere_truncate='END',
+    ),
+)
+```
+
+##### Amazon Nova
+
+- `amazon.nova-2-multimodal-embeddings-v1:0` — Multimodal embeddings, 8K tokens
+
+**Nova-specific settings:**
+
+- `bedrock_nova_truncate` — Truncation strategy: `'NONE'` (default, raises error if input exceeds max tokens), `'START'`, or `'END'`. Overrides the base `truncate` setting if provided.
+- `bedrock_nova_embedding_purpose` — Embedding purpose. Defaults to `'GENERIC_RETRIEVAL'` for `embed_query()` and `'GENERIC_INDEX'` for `embed_documents()`. Other options: `'TEXT_RETRIEVAL'`, `'CLASSIFICATION'`, `'CLUSTERING'`.
+
+```python {title="bedrock_nova.py"}
+from pydantic_ai import Embedder
+from pydantic_ai.embeddings.bedrock import BedrockEmbeddingSettings
+
+embedder = Embedder(
+    'bedrock:amazon.nova-2-multimodal-embeddings-v1:0',
+    settings=BedrockEmbeddingSettings(
+        bedrock_nova_embedding_purpose='TEXT_RETRIEVAL',
+    ),
+)
+```
+
+Nova also supports S3 URIs as input for embedding files stored in S3:
+
+```python {title="bedrock_nova_s3.py" test="skip"}
+from pydantic_ai import Embedder
+
+embedder = Embedder('bedrock:amazon.nova-2-multimodal-embeddings-v1:0')
+
+
+async def main():
+    result = await embedder.embed_query('s3://pydantic-ai-test-bucket/hello-world.txt')
+    print(len(result.embeddings[0]))
+```
+
+#### Regional Prefixes (Cross-Region Inference)
+
+Bedrock supports cross-region inference using geographic prefixes like `us.`, `eu.`, or `apac.`:
+
+```python {title="bedrock_regional.py"}
+from pydantic_ai import Embedder
+
+embedder = Embedder('bedrock:us.amazon.titan-embed-text-v2:0')
+```
+
+!!! note "Token Counting"
+    Bedrock embedding models do not support the `count_tokens()` method because AWS Bedrock's token counting API only works with text generation models (Claude, Llama, etc.), not embedding models. Calling `count_tokens()` will raise `NotImplementedError`.
+
+#### Using a Custom Provider
+
+For advanced configuration like explicit credentials or a custom boto3 client, you can create a [`BedrockProvider`][pydantic_ai.providers.bedrock.BedrockProvider] directly. See the [Bedrock provider documentation](models/bedrock.md#provider-argument) for more details.
+
+```python {title="bedrock_provider.py"}
+from pydantic_ai import Embedder
+from pydantic_ai.embeddings.bedrock import BedrockEmbeddingModel
+from pydantic_ai.providers.bedrock import BedrockProvider
+
+provider = BedrockProvider(
+    region_name='us-west-2',
+    aws_access_key_id='your-access-key',
+    aws_secret_access_key='your-secret-key',
+)
+
+model = BedrockEmbeddingModel('amazon.titan-embed-text-v2:0', provider=provider)
+embedder = Embedder(model)
+```
+
 ### Sentence Transformers (Local)
 
 [`SentenceTransformerEmbeddingModel`][pydantic_ai.embeddings.sentence_transformers.SentenceTransformerEmbeddingModel] runs embeddings locally using the [sentence-transformers](https://www.sbert.net/) library. This is ideal for:

pydantic_ai_slim/pydantic_ai/embeddings/__init__.py

@@ -96,6 +96,10 @@ def infer_embedding_model(
         from .cohere import CohereEmbeddingModel
 
         return CohereEmbeddingModel(model_name, provider=provider)
+    elif model_kind == 'bedrock':
+        from .bedrock import BedrockEmbeddingModel
+
+        return BedrockEmbeddingModel(model_name, provider=provider)
     elif model_kind in ('google-gla', 'google-vertex'):
         from .google import GoogleEmbeddingModel
 

pydantic_ai_slim/pydantic_ai/embeddings/base.py

@@ -14,6 +14,7 @@ class EmbeddingModel(ABC):
     - [`OpenAIEmbeddingModel`][pydantic_ai.embeddings.openai.OpenAIEmbeddingModel]
     - [`CohereEmbeddingModel`][pydantic_ai.embeddings.cohere.CohereEmbeddingModel]
     - [`GoogleEmbeddingModel`][pydantic_ai.embeddings.google.GoogleEmbeddingModel]
+    - [`BedrockEmbeddingModel`][pydantic_ai.embeddings.bedrock.BedrockEmbeddingModel]
     - [`SentenceTransformerEmbeddingModel`][pydantic_ai.embeddings.sentence_transformers.SentenceTransformerEmbeddingModel]
     """