chore(python-sdk): bump version to alpha-6

[teamdandelion]

# Mirascope v2.0.0-alpha.6 Release Notes

## Thinking Support

We've re-worked the thinking interface. Rather than simply setting thinking=True, you now provider a ThinkingConfig, like {"level": "medium", "include_summaries": True}. The ThinkingLevel is required, and options include "default", "none", "minimal", "low", "medium", "high", and "max". If include_summaries is True, then the model is asked to output summaries of its thoughts. Under the hood, Mirascope will convert the thinking level to the provider-specific representation.

## MCP Support

We've added MCP support to Mirascope! You can now connect to an MCP server using functions in the llm.mcp package, and pass the MCP server's tools to your Mirascope call.

Here is a runnable example showing both thinking and MCP. When invoked, Gemini will use the official FastMCP server to learn about FastMCP, and reports back on its findings.

import asyncio

from mirascope import llm

async def main():

    async with llm.mcp.streamable_http_client(
        "https://gofastmcp.com/mcp"
     as mcp_client:

        tools = await mcp_client.list_tools()
        @llm.call(
          "google/gemini-3-flash-preview",
          thinking={"level": "medium", "include_summaries": True},
          tools=tools,
        )

        async def learn_mcp():
            return "Use the tools to learn about FastMCP, and write a report on the library."


        response = await learn_mcp.stream()
        
        while True:  # Loop for tool calls
            async for chunk in response.pretty_stream():
                print(chunk, flush=True, end="")
            print()

            if response.tool_calls:
                tool_output = await response.execute_tools()
                response = await response.resume(tool_output)
            else:
                break

if __name__ == "__main__":
    asyncio.run(main())

## Other Fixes

• Improve OpenAI thinking support to use GPT-5.2 full thinking levels
• Improve Google thinking support to use thinking levels for Gemini-3 models
• Improve Google thinking budget calculation to use a proportion of max tokens
• Refactor llm.Tool construction to make it easier to construct tools that are not decorated functions
• Future-proof streamed tool calls to support providers that stream multiple tool calls simultaneously
• Consolidate formatting utilities onto the llm.Format dataclass
• Improve missing import handling for optional packages (e.g. specific provider sdks)
• Add the llm.MirascopeProvider for future use with Mirascope Router

[teamdandelion]

• chore(python-sdk): bump version to alpha-6 #1863 👈 (View in Graphite)
• chore(python-sdk): update hardcoded model info #1858
• fix(python-sdk): thinking support for gemini-3 #1857
• chore(python-sdk): update google version #1856
• fix(python-sdk): use full range of openai effort levels #1854
• refactor(python-sdk): llm.ThinkingConfig #1847
• v2

This stack of pull requests is managed by Graphite. Learn more about stacking.

[cloudflare-workers-and-pages]

Deploying with    Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
✅ Deployment successful! View logs	v2-docs	efaea43	Jan 10 2026, 01:57 AM

[willbakst]

Merge activity

• Jan 10, 1:53 AM UTC: A user started a stack merge that includes this pull request via Graphite.
• Jan 10, 1:53 AM UTC: Graphite rebased this pull request as part of a merge.
• Jan 10, 1:58 AM UTC: @willbakst merged this pull request with Graphite.