chore: merge main

.github/workflows/lint.yml

@@ -5,9 +5,11 @@ on:
   push:
     branches:
       - main
+      - release/*
   pull_request:
     branches:
       - main
+      - release/*
 
 jobs:
   lints:

.github/workflows/tests.yml

@@ -5,9 +5,11 @@ on:
   push:
     branches:
       - main
+      - release/*
   pull_request:
     branches:
       - main
+      - release/*
 
 jobs:
   tests:

README.md

@@ -16,7 +16,7 @@
 
 ---
 
-Mirascope is a powerful, flexible, and user-friendly library that simplifies the process of working with LLMs through a unified interface that works across various supported providers, including [OpenAI](https://openai.com/), [Anthropic](https://www.anthropic.com/), [Mistral](https://mistral.ai/), [Gemini](https://gemini.google.com), [Groq](https://groq.com/), [Cohere](https://cohere.com/), [LiteLLM](https://www.litellm.ai/), [Azure AI](https://azure.microsoft.com/en-us/solutions/ai), and [Vertex AI](https://cloud.google.com/vertex-ai).
+Mirascope is a powerful, flexible, and user-friendly library that simplifies the process of working with LLMs through a unified interface that works across various supported providers, including [OpenAI](https://openai.com/), [Anthropic](https://www.anthropic.com/), [Mistral](https://mistral.ai/), [Gemini](https://gemini.google.com), [Groq](https://groq.com/), [Cohere](https://cohere.com/), [LiteLLM](https://www.litellm.ai/), [Azure AI](https://azure.microsoft.com/en-us/solutions/ai), [Vertex AI](https://cloud.google.com/vertex-ai), and [Bedrock](https://aws.amazon.com/bedrock/).
 
 Whether you're generating text, extracting structured information, or developing complex AI-driven agent systems, Mirascope provides the tools you need to streamline your development process and create powerful, robust applications.
 

docs/WHY.md

@@ -39,6 +39,8 @@ Let's compare structured outputs using Mirascope vs. the official SDKs:
         ```python hl_lines="26-46 51"
         {% elif provider == "Vertex AI" %}
         ```python hl_lines="23-62 67"
+        {% elif provider == "Bedrock" %}
+        ```python hl_lines="17-48 53"
         {% else %}
         ```python hl_lines="18-39 44"
         {% endif %}

docs/api/core/bedrock/call.md

@@ -0,0 +1,3 @@
+# mirascope.core.bedrock.call
+
+::: mirascope.core.bedrock.call

docs/api/core/bedrock/call_params.md

@@ -0,0 +1,3 @@
+# mirascope.core.bedrock.call_params
+
+::: mirascope.core.bedrock.call_params

docs/api/core/bedrock/call_response.md

@@ -0,0 +1,3 @@
+# mirascope.core.bedrock.call_response
+
+::: mirascope.core.bedrock.call_response

docs/api/core/bedrock/call_response_chunk.md

@@ -0,0 +1,3 @@
+# mirascope.core.bedrock.call_response_chunk
+
+::: mirascope.core.bedrock.call_response_chunk

docs/api/core/bedrock/dynamic_config.md

@@ -0,0 +1,3 @@
+# mirascope.core.bedrock.dynamic_config
+
+::: mirascope.core.bedrock.dynamic_config

docs/api/core/bedrock/stream.md

@@ -0,0 +1,3 @@
+# mirascope.core.bedrock.stream
+
+::: mirascope.core.bedrock.stream

docs/api/core/bedrock/tool.md

@@ -0,0 +1,3 @@
+# mirascope.core.bedrock.tool
+
+::: mirascope.core.bedrock.tool

docs/extra/tweaks.css

@@ -1,14 +1,16 @@
+/* Color Palette */
+:root {
+  --md-primary-fg-color:        #6366f1;
+  --md-primary-fg-color--light: #6366f1;
+  --md-primary-fg-color--dark:  #6366f1;
+}
+
 /* 
  * Original credit for this admonition implementation goes to the team at Pydantic
  * https://github.com/pydantic/pydantic/blob/main/docs/extra/tweaks.css
 */
 
 /* API documentation link admonition */
-:root {
-  --md-primary-fg-color:        #6366f1;
-  --md-primary-fg-color--light: #6366f1;
-  --md-primary-fg-color--dark:  #6366f1;
-}
 
 :root {
     --md-admonition-icon--api: url('data:image/svg+xml;charset=utf-8,<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M7 7H5a2 2 0 0 0-2 2v8h2v-4h2v4h2V9a2 2 0 0 0-2-2m0 4H5V9h2m7-2h-4v10h2v-4h2a2 2 0 0 0 2-2V9a2 2 0 0 0-2-2m0 4h-2V9h2m6 0v6h1v2h-4v-2h1V9h-1V7h4v2Z"/></svg>')

docs/index.md

@@ -11,7 +11,7 @@
 
 LLM abstractions that aren't obstructions.
 
-Mirascope is a powerful, flexible, and user-friendly library that simplifies the process of working with LLMs through a unified interface that works across various supported providers, including [OpenAI](https://openai.com/), [Anthropic](https://www.anthropic.com/), [Mistral](https://mistral.ai/), [Gemini](https://gemini.google.com), [Groq](https://groq.com/), [Cohere](https://cohere.com/), [LiteLLM](https://www.litellm.ai/), [Azure AI](https://azure.microsoft.com/en-us/solutions/ai), and [Vertex AI](https://cloud.google.com/vertex-ai).
+Mirascope is a powerful, flexible, and user-friendly library that simplifies the process of working with LLMs through a unified interface that works across various supported providers, including [OpenAI](https://openai.com/), [Anthropic](https://www.anthropic.com/), [Mistral](https://mistral.ai/), [Gemini](https://gemini.google.com), [Groq](https://groq.com/), [Cohere](https://cohere.com/), [LiteLLM](https://www.litellm.ai/), [Azure AI](https://azure.microsoft.com/en-us/solutions/ai), [Vertex AI](https://cloud.google.com/vertex-ai), and [Bedrock](https://aws.amazon.com/bedrock/).
 
 Whether you're generating text, extracting structured information, or developing complex AI-driven agent systems, Mirascope provides the tools you need to streamline your development process and create powerful, robust applications.
 
@@ -49,6 +49,8 @@ Install Mirascope, specifying the provider(s) you intend to use, and set your AP
         {% elif provider == "Vertex AI" %}
         gcloud init
         gcloud auth application-default login
+        {% elif provider == "Bedrock" %}
+        aws configure
         {% else %}
         {% if os == "Windows" %}set {{ upper(provider | provider_dir) }}_API_KEY=XXXXX
         {% else %}export {{ upper(provider | provider_dir) }}_API_KEY=XXXXX

docs/learn/async.md

@@ -175,6 +175,8 @@ It's important to note that you must use the correct client that supports asynch
             ```python hl_lines="2 5"
             {% elif provider == "Azure AI" %}
             ```python hl_lines="1-2 8-10"
+            {% elif provider == "Bedrock" %}
+            ```python hl_lines="7-10 14"
             {% else %}
             ```python hl_lines="1 5"
             {% endif %}

docs/learn/calls.md

@@ -303,6 +303,8 @@ To use a custom client, you can pass it to the `call` decorator using the `clien
             ```python hl_lines="2 5"
             {% elif provider == "Azure AI" %}
             ```python hl_lines="1-2 8-10"
+            {% elif provider == "Bedrock" %}
+            ```python hl_lines="1 6"
             {% else %}
             ```python hl_lines="1 5"
             {% endif %}

docs/learn/output_parsers.md

@@ -27,12 +27,19 @@ Let's take a look at a basic example:
 
         {% for provider in supported_llm_providers %}
         === "{{ provider }}"
-
             {% if method == "base_message_param" %}
+            {% if provider == "Bedrock" %}
+            ```python hl_lines="10 22"
+            {% else %}
             ```python hl_lines="9 20"
+            {% endif %}
+            {% else %}
+            {% if provider == "Bedrock" %} 
+            ```python hl_lines="10 17"
             {% else %}
             ```python hl_lines="9 15"
             {% endif %}
+            {% endif %}
             --8<-- "examples/learn/output_parsers/basic_usage/{{ provider | provider_dir }}/{{ method }}.py"
             ```
         {% endfor %}

docs/learn/response_models.md

@@ -190,8 +190,11 @@ By default, `response_model` uses [Tools](./tools.md) under the hood. You can in
 
         {% for provider in supported_llm_providers %}
         === "{{ provider }}"
-
+            {% if provider == "Bedrock" %}
+            ```python hl_lines="13"
+            {% else %}
             ```python hl_lines="12"
+            {% endif %}
             --8<-- "examples/learn/response_models/json_mode/{{ provider | provider_dir }}/{{ method }}.py"
             ```
         {% endfor %}
@@ -238,8 +241,11 @@ If you set `stream=True` when `response_model` is set, your LLM call will return
 
         {% for provider in supported_llm_providers %}
         === "{{ provider }}"
-
+            {% if provider == "Bedrock" %}
+            ```python hl_lines="11 18-19"
+            {% else %}
             ```python hl_lines="10 16-17"
+            {% endif %}
             --8<-- "examples/learn/response_models/streaming/{{ provider | provider_dir }}/{{ method }}.py"
             ```
         {% endfor %}
@@ -250,6 +256,29 @@ Once exhausted, you can access the final, full response model through the `const
 
 You can also use the `stream` property to access the `BaseStream` instance and [all of it's properties](./streams.md#common-stream-properties-and-methods).
 
+## FromCallArgs
+
+Fields annotated with `FromCallArgs` will be populated with the corresponding argument from the function call rather than expecting it from the LLM's response. This enables seamless validation of LLM outputs against function inputs:
+
+!!! mira "Mirascope"
+
+    {% for method, method_title in zip(prompt_writing_methods, prompt_writing_method_titles) %}
+    === "{{ method_title }}"
+
+        {% for provider in supported_llm_providers %}
+        === "{{ provider }}"
+
+            {% if method == "string_template" %}
+            ```python hl_lines="14 26"
+            {% else %}
+            ```python hl_lines="14 25"
+            {% endif %}
+            --8<-- "examples/learn/response_models/from_call_args/{{ provider | provider_dir }}/{{ method }}.py"
+            ```
+        {% endfor %}
+
+    {% endfor %}
+
 ## Next Steps
 
 By following these best practices and leveraging Response Models effectively, you can create more robust, type-safe, and maintainable LLM-powered applications with Mirascope.

docs/learn/tools.md

@@ -223,10 +223,18 @@ Mirascope supports streaming responses with tools, which is useful for long-runn
             === "{{ provider }}"
 
                 {% if tool_method == "function" %}
+                {% if provider == "Bedrock" %}
+                ```python hl_lines="19 26-28"
+                {% else %}
                 ```python hl_lines="18 24-26"
+                {% endif %}
+                {% else %}
+                {% if provider == "Bedrock" %}
+                ```python hl_lines="20 27-29"
                 {% else %}
                 ```python hl_lines="19 25-27"
                 {% endif %}
+                {% endif %}
                 --8<-- "examples/learn/tools/streams/{{ provider | provider_dir }}/{{ tool_method }}/{{ method }}.py"
                 ```
             {% endfor %}