github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-21 06:14:37 +00:00
Code Issues Packages Projects Releases Wiki Activity

docs: update Bedrock chat model page (#30883)

Browse Source 
- document prompt caching
- feature ChatBedrockConverse throughout
This commit is contained in:
ccurme 2025-04-16 16:55:14 -04:00 committed by GitHub
parent 9e82f1df4e
commit b36c2bf833
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 409 additions and 377 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

315
docs/docs/integrations/chat/bedrock.ipynb
View File

@ -19,9 +19,15 @@
"\n", "\n",
"This doc will help you get started with AWS Bedrock [chat models](/docs/concepts/chat_models). Amazon Bedrock is a fully managed service that offers a choice of high-performing foundation models (FMs) from leading AI companies like AI21 Labs, Anthropic, Cohere, Meta, Stability AI, and Amazon via a single API, along with a broad set of capabilities you need to build generative AI applications with security, privacy, and responsible AI. Using Amazon Bedrock, you can easily experiment with and evaluate top FMs for your use case, privately customize them with your data using techniques such as fine-tuning and Retrieval Augmented Generation (RAG), and build agents that execute tasks using your enterprise systems and data sources. Since Amazon Bedrock is serverless, you don't have to manage any infrastructure, and you can securely integrate and deploy generative AI capabilities into your applications using the AWS services you are already familiar with.\n", "This doc will help you get started with AWS Bedrock [chat models](/docs/concepts/chat_models). Amazon Bedrock is a fully managed service that offers a choice of high-performing foundation models (FMs) from leading AI companies like AI21 Labs, Anthropic, Cohere, Meta, Stability AI, and Amazon via a single API, along with a broad set of capabilities you need to build generative AI applications with security, privacy, and responsible AI. Using Amazon Bedrock, you can easily experiment with and evaluate top FMs for your use case, privately customize them with your data using techniques such as fine-tuning and Retrieval Augmented Generation (RAG), and build agents that execute tasks using your enterprise systems and data sources. Since Amazon Bedrock is serverless, you don't have to manage any infrastructure, and you can securely integrate and deploy generative AI capabilities into your applications using the AWS services you are already familiar with.\n",
"\n", "\n",
"For more information on which models are accessible via Bedrock, head to the [AWS docs](https://docs.aws.amazon.com/bedrock/latest/userguide/models-features.html).\n", "AWS Bedrock maintains a [Converse API](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_Converse.html) which provides a unified conversational interface for Bedrock models. This API does not yet support custom models. You can see a list of all [models that are supported here](https://docs.aws.amazon.com/bedrock/latest/userguide/conversation-inference.html).\n",
"\n", "\n",
"For detailed documentation of all ChatBedrock features and configurations head to the [API reference](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock.ChatBedrock.html).\n", ":::info\n",
"\n",
"We recommend the Converse API for users who do not need to use custom models. It can be accessed using [ChatBedrockConverse](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock_converse.ChatBedrockConverse.html).\n",
"\n",
":::\n",
"\n",
"For detailed documentation of all Bedrock features and configurations head to the [API reference](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock_converse.ChatBedrockConverse.html).\n",
"\n", "\n",
"## Overview\n", "## Overview\n",
"### Integration details\n", "### Integration details\n",
@ -29,8 +35,12 @@
"| Class | Package | Local | Serializable | [JS support](https://js.langchain.com/docs/integrations/chat/bedrock) | Package downloads | Package latest |\n", "| Class | Package | Local | Serializable | [JS support](https://js.langchain.com/docs/integrations/chat/bedrock) | Package downloads | Package latest |\n",
"| :--- | :--- | :---: | :---: | :---: | :---: | :---: |\n", "| :--- | :--- | :---: | :---: | :---: | :---: | :---: |\n",
"| [ChatBedrock](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock.ChatBedrock.html) | [langchain-aws](https://python.langchain.com/api_reference/aws/index.html) | ❌ | beta | ✅ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-aws?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-aws?style=flat-square&label=%20) |\n", "| [ChatBedrock](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock.ChatBedrock.html) | [langchain-aws](https://python.langchain.com/api_reference/aws/index.html) | ❌ | beta | ✅ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-aws?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-aws?style=flat-square&label=%20) |\n",
"| [ChatBedrockConverse](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock_converse.ChatBedrockConverse.html) | [langchain-aws](https://python.langchain.com/api_reference/aws/index.html) | ❌ | beta | ✅ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-aws?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-aws?style=flat-square&label=%20) |\n",
"\n", "\n",
"### Model features\n", "### Model features\n",
"\n",
"The below apply to both `ChatBedrock` and `ChatBedrockConverse`.\n",
"\n",
"| [Tool calling](/docs/how_to/tool_calling) | [Structured output](/docs/how_to/structured_output/) | JSON mode | [Image input](/docs/how_to/multimodal_inputs/) | Audio input | Video input | [Token-level streaming](/docs/how_to/chat_streaming/) | Native async | [Token usage](/docs/how_to/chat_token_usage_tracking/) | [Logprobs](/docs/how_to/logprobs/) |\n", "| [Tool calling](/docs/how_to/tool_calling) | [Structured output](/docs/how_to/structured_output/) | JSON mode | [Image input](/docs/how_to/multimodal_inputs/) | Audio input | Video input | [Token-level streaming](/docs/how_to/chat_streaming/) | Native async | [Token usage](/docs/how_to/chat_token_usage_tracking/) | [Logprobs](/docs/how_to/logprobs/) |\n",
"| :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |\n", "| :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |\n",
"| ✅ | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ❌ | ✅ | ❌ |\n", "| ✅ | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ❌ | ✅ | ❌ |\n",
@ -48,7 +58,9 @@
"cell_type": "markdown", "cell_type": "markdown",
"id": "72ee0c4b-9764-423a-9dbf-95129e185210", "id": "72ee0c4b-9764-423a-9dbf-95129e185210",
"metadata": {}, "metadata": {},
"source": "To enable automated tracing of your model calls, set your [LangSmith](https://docs.smith.langchain.com/) API key:" "source": [
"To enable automated tracing of your model calls, set your [LangSmith](https://docs.smith.langchain.com/) API key:"
]
}, },
{ {
"cell_type": "code", "cell_type": "code",
@ -98,11 +110,12 @@
"metadata": {}, "metadata": {},
"outputs": [], "outputs": [],
"source": [ "source": [
"from langchain_aws import ChatBedrock\n", "from langchain_aws import ChatBedrockConverse\n",
"\n", "\n",
"llm = ChatBedrock(\n", "llm = ChatBedrockConverse(\n",
" model_id=\"anthropic.claude-3-sonnet-20240229-v1:0\",\n", " model_id=\"anthropic.claude-3-5-sonnet-20240620-v1:0\",\n",
" model_kwargs=dict(temperature=0),\n", " # temperature=...,\n",
" # max_tokens=...,\n",
" # other params...\n", " # other params...\n",
")" ")"
] ]
@ -117,19 +130,17 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 5, "execution_count": 2,
"id": "62e0dbc3", "id": "fcd8de52-4a1b-4875-b463-d41b031e06a1",
"metadata": { "metadata": {},
"tags": []
},
"outputs": [ "outputs": [
{ {
"data": { "data": {
"text/plain": [ "text/plain": [
"AIMessage(content=\"Voici la traduction en français :\\n\\nJ'aime la programmation.\", additional_kwargs={'usage': {'prompt_tokens': 29, 'completion_tokens': 21, 'total_tokens': 50}, 'stop_reason': 'end_turn', 'model_id': 'anthropic.claude-3-sonnet-20240229-v1:0'}, response_metadata={'usage': {'prompt_tokens': 29, 'completion_tokens': 21, 'total_tokens': 50}, 'stop_reason': 'end_turn', 'model_id': 'anthropic.claude-3-sonnet-20240229-v1:0'}, id='run-fdb07dc3-ff72-430d-b22b-e7824b15c766-0', usage_metadata={'input_tokens': 29, 'output_tokens': 21, 'total_tokens': 50})" "AIMessage(content=\"J'adore la programmation.\", additional_kwargs={}, response_metadata={'ResponseMetadata': {'RequestId': 'b07d1630-06f2-44b1-82bf-e82538dd2215', 'HTTPStatusCode': 200, 'HTTPHeaders': {'date': 'Wed, 16 Apr 2025 19:35:34 GMT', 'content-type': 'application/json', 'content-length': '206', 'connection': 'keep-alive', 'x-amzn-requestid': 'b07d1630-06f2-44b1-82bf-e82538dd2215'}, 'RetryAttempts': 0}, 'stopReason': 'end_turn', 'metrics': {'latencyMs': [488]}, 'model_name': 'anthropic.claude-3-5-sonnet-20240620-v1:0'}, id='run-d09ed928-146a-4336-b1fd-b63c9e623494-0', usage_metadata={'input_tokens': 29, 'output_tokens': 11, 'total_tokens': 40, 'input_token_details': {'cache_creation': 0, 'cache_read': 0}})"
] ]
}, },
"execution_count": 5, "execution_count": 2,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
} }
@ -148,7 +159,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 6, "execution_count": 3,
"id": "d86145b3-bfef-46e8-b227-4dda5c9c2705", "id": "d86145b3-bfef-46e8-b227-4dda5c9c2705",
"metadata": {}, "metadata": {},
"outputs": [ "outputs": [
@ -156,9 +167,7 @@
"name": "stdout", "name": "stdout",
"output_type": "stream", "output_type": "stream",
"text": [ "text": [
"Voici la traduction en français :\n", "J'adore la programmation.\n"
"\n",
"J'aime la programmation.\n"
] ]
} }
], ],
@ -168,7 +177,146 @@
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
"id": "18e2bfc0-7e78-4528-a73f-499ac150dca8", "id": "4da16f3e-e80b-48c0-8036-c1cc5f7c8c05",
"metadata": {},
"source": [
"### Streaming\n",
"\n",
"Note that `ChatBedrockConverse` emits content blocks while streaming:"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "605e04fa-1a76-47ac-8c92-fe128659663e",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"content=[] additional_kwargs={} response_metadata={} id='run-d0e0836e-7146-4c3d-97c7-ad23dac6febd'\n",
"content=[{'type': 'text', 'text': 'J', 'index': 0}] additional_kwargs={} response_metadata={} id='run-d0e0836e-7146-4c3d-97c7-ad23dac6febd'\n",
"content=[{'type': 'text', 'text': \"'adore la\", 'index': 0}] additional_kwargs={} response_metadata={} id='run-d0e0836e-7146-4c3d-97c7-ad23dac6febd'\n",
"content=[{'type': 'text', 'text': ' programmation.', 'index': 0}] additional_kwargs={} response_metadata={} id='run-d0e0836e-7146-4c3d-97c7-ad23dac6febd'\n",
"content=[{'index': 0}] additional_kwargs={} response_metadata={} id='run-d0e0836e-7146-4c3d-97c7-ad23dac6febd'\n",
"content=[] additional_kwargs={} response_metadata={'stopReason': 'end_turn'} id='run-d0e0836e-7146-4c3d-97c7-ad23dac6febd'\n",
"content=[] additional_kwargs={} response_metadata={'metrics': {'latencyMs': 600}, 'model_name': 'anthropic.claude-3-5-sonnet-20240620-v1:0'} id='run-d0e0836e-7146-4c3d-97c7-ad23dac6febd' usage_metadata={'input_tokens': 29, 'output_tokens': 11, 'total_tokens': 40, 'input_token_details': {'cache_creation': 0, 'cache_read': 0}}\n"
]
}
],
"source": [
"for chunk in llm.stream(messages):\n",
" print(chunk)"
]
},
{
"cell_type": "markdown",
"id": "0ef05abb-9c04-4dc3-995e-f857779644d5",
"metadata": {},
"source": [
"You can filter to text using the [.text()](https://python.langchain.com/api_reference/core/messages/langchain_core.messages.ai.AIMessage.html#langchain_core.messages.ai.AIMessage.text) method on the output:"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "2a4e743f-ea7d-4e5a-9b12-f9992362de8b",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"|J|'adore la| programmation.||||"
]
}
],
"source": [
"for chunk in llm.stream(messages):\n",
" print(chunk.text(), end=\"|\")"
]
},
{
"cell_type": "markdown",
"id": "a77519e5-897d-41a0-a9bb-55300fa79efc",
"metadata": {},
"source": [
"## Prompt caching\n",
"\n",
"Bedrock supports [caching](https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-caching.html) of elements of your prompts, including messages and tools. This allows you to re-use large documents, instructions, [few-shot documents](/docs/concepts/few_shot_prompting/), and other data to reduce latency and costs.\n",
"\n",
":::note\n",
"\n",
"Not all models support prompt caching. See supported models [here](https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-caching.html#prompt-caching-models).\n",
"\n",
":::\n",
"\n",
"To enable caching on an element of a prompt, mark its associated content block using the `cachePoint` key. See example below:"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "d5f63d01-85e8-4797-a2be-0fea747a6049",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"First invocation:\n",
"{'cache_creation': 1528, 'cache_read': 0}\n",
"\n",
"Second:\n",
"{'cache_creation': 0, 'cache_read': 1528}\n"
]
}
],
"source": [
"import requests\n",
"from langchain_aws import ChatBedrockConverse\n",
"\n",
"llm = ChatBedrockConverse(model=\"us.anthropic.claude-3-7-sonnet-20250219-v1:0\")\n",
"\n",
"# Pull LangChain readme\n",
"get_response = requests.get(\n",
" \"https://raw.githubusercontent.com/langchain-ai/langchain/master/README.md\"\n",
")\n",
"readme = get_response.text\n",
"\n",
"messages = [\n",
" {\n",
" \"role\": \"user\",\n",
" \"content\": [\n",
" {\n",
" \"type\": \"text\",\n",
" \"text\": \"What's LangChain, according to its README?\",\n",
" },\n",
" {\n",
" \"type\": \"text\",\n",
" \"text\": f\"{readme}\",\n",
" },\n",
" {\n",
" \"cachePoint\": {\"type\": \"default\"},\n",
" },\n",
" ],\n",
" },\n",
"]\n",
"\n",
"response_1 = llm.invoke(messages)\n",
"response_2 = llm.invoke(messages)\n",
"\n",
"usage_1 = response_1.usage_metadata[\"input_token_details\"]\n",
"usage_2 = response_2.usage_metadata[\"input_token_details\"]\n",
"\n",
"print(f\"First invocation:\\n{usage_1}\")\n",
"print(f\"\\nSecond:\\n{usage_2}\")"
]
},
{
"cell_type": "markdown",
"id": "1b550667-af5b-4557-b84f-c8f865dad6cb",
"metadata": {}, "metadata": {},
"source": [ "source": [
"## Chaining\n", "## Chaining\n",
@ -179,13 +327,13 @@
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 7, "execution_count": 7,
"id": "e197d1d7-a070-4c96-9f8a-a0e86d046e0b", "id": "6033f3fa-0e96-46e3-abb3-1530928fea88",
"metadata": {}, "metadata": {},
"outputs": [ "outputs": [
{ {
"data": { "data": {
"text/plain": [ "text/plain": [
"AIMessage(content='Ich liebe Programmieren.', additional_kwargs={'usage': {'prompt_tokens': 23, 'completion_tokens': 11, 'total_tokens': 34}, 'stop_reason': 'end_turn', 'model_id': 'anthropic.claude-3-sonnet-20240229-v1:0'}, response_metadata={'usage': {'prompt_tokens': 23, 'completion_tokens': 11, 'total_tokens': 34}, 'stop_reason': 'end_turn', 'model_id': 'anthropic.claude-3-sonnet-20240229-v1:0'}, id='run-5ad005ce-9f31-4670-baa0-9373d418698a-0', usage_metadata={'input_tokens': 23, 'output_tokens': 11, 'total_tokens': 34})" "AIMessage(content=\"Here's the German translation:\\n\\nIch liebe das Programmieren.\", additional_kwargs={}, response_metadata={'ResponseMetadata': {'RequestId': '1de3d7c0-8062-4f7e-bb8a-8f725b97a8b0', 'HTTPStatusCode': 200, 'HTTPHeaders': {'date': 'Wed, 16 Apr 2025 19:32:51 GMT', 'content-type': 'application/json', 'content-length': '243', 'connection': 'keep-alive', 'x-amzn-requestid': '1de3d7c0-8062-4f7e-bb8a-8f725b97a8b0'}, 'RetryAttempts': 0}, 'stopReason': 'end_turn', 'metrics': {'latencyMs': [719]}, 'model_name': 'anthropic.claude-3-5-sonnet-20240620-v1:0'}, id='run-7021fcd7-704e-496b-a92e-210139614402-0', usage_metadata={'input_tokens': 23, 'output_tokens': 19, 'total_tokens': 42, 'input_token_details': {'cache_creation': 0, 'cache_read': 0}})"
] ]
}, },
"execution_count": 7, "execution_count": 7,
@ -216,131 +364,6 @@
")" ")"
] ]
}, },
{
"cell_type": "markdown",
"id": "d1ee55bc-ffc8-4cfa-801c-993953a08cfd",
"metadata": {},
"source": [
"## Bedrock Converse API\n",
"\n",
"AWS has recently released the Bedrock Converse API which provides a unified conversational interface for Bedrock models. This API does not yet support custom models. You can see a list of all [models that are supported here](https://docs.aws.amazon.com/bedrock/latest/userguide/conversation-inference.html). To improve reliability the ChatBedrock integration will switch to using the Bedrock Converse API as soon as it has feature parity with the existing Bedrock API. Until then a separate [ChatBedrockConverse](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock_converse.ChatBedrockConverse.html) integration has been released.\n",
"\n",
"We recommend using `ChatBedrockConverse` for users who do not need to use custom models.\n",
"\n",
"You can use it like so:"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "ae728e59-94d4-40cf-9d24-25ad8723fc59",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"AIMessage(content=\"Voici la traduction en français :\\n\\nJ'aime la programmation.\", response_metadata={'ResponseMetadata': {'RequestId': '4fcbfbe9-f916-4df2-b0bd-ea1147b550aa', 'HTTPStatusCode': 200, 'HTTPHeaders': {'date': 'Wed, 21 Aug 2024 17:23:49 GMT', 'content-type': 'application/json', 'content-length': '243', 'connection': 'keep-alive', 'x-amzn-requestid': '4fcbfbe9-f916-4df2-b0bd-ea1147b550aa'}, 'RetryAttempts': 0}, 'stopReason': 'end_turn', 'metrics': {'latencyMs': 672}}, id='run-77ee9810-e32b-45dc-9ccb-6692253b1f45-0', usage_metadata={'input_tokens': 29, 'output_tokens': 21, 'total_tokens': 50})"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from langchain_aws import ChatBedrockConverse\n",
"\n",
"llm = ChatBedrockConverse(\n",
" model=\"anthropic.claude-3-sonnet-20240229-v1:0\",\n",
" temperature=0,\n",
" max_tokens=None,\n",
" # other params...\n",
")\n",
"\n",
"llm.invoke(messages)"
]
},
{
"cell_type": "markdown",
"id": "4da16f3e-e80b-48c0-8036-c1cc5f7c8c05",
"metadata": {},
"source": [
"### Streaming\n",
"\n",
"Note that `ChatBedrockConverse` emits content blocks while streaming:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "7794b32e-d8de-4973-bf0f-39807dc745f0",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"content=[] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': 'Vo', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': 'ici', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': ' la', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': ' tra', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': 'duction', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': ' en', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': ' français', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': ' :', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': '\\n\\nJ', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': \"'\", 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': 'a', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': 'ime', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': ' la', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': ' programm', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': 'ation', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'type': 'text', 'text': '.', 'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[{'index': 0}] id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[] response_metadata={'stopReason': 'end_turn'} id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8'\n",
"content=[] response_metadata={'metrics': {'latencyMs': 713}} id='run-2c92c5af-d771-4cc2-98d9-c11bbd30a1d8' usage_metadata={'input_tokens': 29, 'output_tokens': 21, 'total_tokens': 50}\n"
]
}
],
"source": [
"for chunk in llm.stream(messages):\n",
" print(chunk)"
]
},
{
"cell_type": "markdown",
"id": "0ef05abb-9c04-4dc3-995e-f857779644d5",
"metadata": {},
"source": [
"An output parser can be used to filter to text, if desired:"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "2a4e743f-ea7d-4e5a-9b12-f9992362de8b",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"|Vo|ici| la| tra|duction| en| français| :|\n",
"\n",
"J|'|a|ime| la| programm|ation|.||||"
]
}
],
"source": [
"from langchain_core.output_parsers import StrOutputParser\n",
"\n",
"chain = llm | StrOutputParser()\n",
"\n",
"for chunk in chain.stream(messages):\n",
" print(chunk, end=\"|\")"
]
},
{ {
"cell_type": "markdown", "cell_type": "markdown",
"id": "3a5bb5ca-c3ae-4a58-be67-2cd18574b9a3", "id": "3a5bb5ca-c3ae-4a58-be67-2cd18574b9a3",

13
docs/docs/integrations/providers/aws.mdx
View File

@ -35,9 +35,18 @@ from langchain_aws import ChatBedrock
``` ```
### Bedrock Converse ### Bedrock Converse
AWS has recently released the Bedrock Converse API which provides a unified conversational interface for Bedrock models. This API does not yet support custom models. You can see a list of all [models that are supported here](https://docs.aws.amazon.com/bedrock/latest/userguide/conversation-inference.html). To improve reliability the ChatBedrock integration will switch to using the Bedrock Converse API as soon as it has feature parity with the existing Bedrock API. Until then a separate [ChatBedrockConverse](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock_converse.ChatBedrockConverse.html) integration has been released. AWS Bedrock maintains a [Converse API](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_Converse.html)
that provides a unified conversational interface for Bedrock models. This API does not
yet support custom models. You can see a list of all
[models that are supported here](https://docs.aws.amazon.com/bedrock/latest/userguide/conversation-inference.html).
We recommend using `ChatBedrockConverse` for users who do not need to use custom models. See the [docs](/docs/integrations/chat/bedrock/#bedrock-converse-api) and [API reference](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock_converse.ChatBedrockConverse.html) for more detail. :::info
We recommend the Converse API for users who do not need to use custom models. It can be accessed using [ChatBedrockConverse](https://python.langchain.com/api_reference/aws/chat_models/langchain_aws.chat_models.bedrock_converse.ChatBedrockConverse.html).
:::
See a [usage example](/docs/integrations/chat/bedrock).
```python ```python
from langchain_aws import ChatBedrockConverse from langchain_aws import ChatBedrockConverse