add mimo example with openrouter

[maxkahan]

Summary by CodeRabbit

• Bug Fixes

  • Fixed streaming response handling to properly terminate after completion signal instead of continuing to consume additional chunks.

• New Features

  • Added example implementation for MiMo Video Assistant using the MiMo-V2-Omni model through OpenRouter, including speech-to-text and text-to-speech capabilities.

[coderabbitai]

📝 Walkthrough

Walkthrough

A streaming response handling fix terminates the completion loop upon encountering finish_reason, preventing further chunk consumption. Additionally, a new example script demonstrates configuring a MiMo-V2-Omni video assistant via OpenRouter with speech and vision capabilities.

Changes

Cohort / File(s)	Summary
Stream Termination Fix plugins/openai/.../chat_completions_vlm.py	Modified streaming loop to assign final llm_response and break immediately after finish_reason is encountered, preventing additional chunk iteration before response return.
MiMo Video Assistant Example plugins/openrouter/example/mimo_example.py	New example script configuring an asynchronous video assistant using Xiaomi's MiMo-V2-Omni model through OpenRouter, including agent creation with GetStream connectivity and speech capabilities.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~22 minutes

Poem

A stream once bled without a name,
chunks cascading past the end—
now silence comes when finish calls,
the loop knows when to break,
and MiMo sees, at last, the light.

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main addition in the changeset: a new MiMo example using OpenRouter, which is the primary focus of the PR.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch add-mimo-example

📝 Coding Plan

• Generate coding plan for human review comments

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

Tip

You can get early access to new features in CodeRabbit.

Enable the early_access setting to enable early access features such as new models, tools, and more.

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

plugins/openrouter/example/mimo_example.py (1)

31-38: Consider graceful handling when OPENROUTER_API_KEY is absent.

The direct os.environ["OPENROUTER_API_KEY"] access will raise a bare KeyError if the variable is unset—a stark, unhelpful death. For better user experience, consider validation with a descriptive error:

🛡️ Optional: Add explicit validation

 async def create_agent(**kwargs) -> Agent:
     """Create a video assistant powered by Xiaomi MiMo-V2-Omni."""
+    api_key = os.environ.get("OPENROUTER_API_KEY")
+    if not api_key:
+        raise ValueError("OPENROUTER_API_KEY environment variable is required")
     llm = openai.ChatCompletionsVLM(
         model="xiaomi/mimo-v2-omni",
         base_url="https://openrouter.ai/api/v1",
-        api_key=os.environ["OPENROUTER_API_KEY"],
+        api_key=api_key,
         frame_buffer_seconds=3,
         frame_width=512,
         frame_height=384,
     )

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@plugins/openrouter/example/mimo_example.py` around lines 31 - 38, The code
directly uses os.environ["OPENROUTER_API_KEY"] when constructing
openai.ChatCompletionsVLM (assigned to llm), which raises KeyError if the env
var is missing; change to validate the API key first (e.g., use
os.getenv("OPENROUTER_API_KEY") or os.environ.get and check for None/empty) and
raise or log a clear, descriptive error before calling openai.ChatCompletionsVLM
so the user sees a helpful message rather than a bare KeyError.

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In
`@plugins/openai/vision_agents/plugins/openai/chat_completions/chat_completions_vlm.py`:
- Around line 269-271: The stream loop currently creates
LLMResponseEvent(original=chunk, text=total_text) and immediately uses break,
which prevents draining any subsequent usage-only chunk and so omits
input/output token metadata; instead, stop breaking on first finish_reason —
continue consuming the generator until it naturally ends, preserve the final
chunk (e.g., final_chunk variable or update the last seen chunk) and after the
loop call the existing _extract_usage_tokens(final_chunk) to populate usage
fields (input_tokens/output_tokens) on the LLMResponseEvent before returning;
update logic around LLMResponseEvent creation and the break in
chat_completions_vlm.py to mirror Gemini’s pattern so
stream_options={"include_usage": True} is honored.

---

Nitpick comments:
In `@plugins/openrouter/example/mimo_example.py`:
- Around line 31-38: The code directly uses os.environ["OPENROUTER_API_KEY"]
when constructing openai.ChatCompletionsVLM (assigned to llm), which raises
KeyError if the env var is missing; change to validate the API key first (e.g.,
use os.getenv("OPENROUTER_API_KEY") or os.environ.get and check for None/empty)
and raise or log a clear, descriptive error before calling
openai.ChatCompletionsVLM so the user sees a helpful message rather than a bare
KeyError.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 45c20e41-6100-429e-8362-4eb97560f177

📥 Commits

Reviewing files that changed from the base of the PR and between b775230 and d3219bb.

📒 Files selected for processing (2)

• plugins/openai/vision_agents/plugins/openai/chat_completions/chat_completions_vlm.py
• plugins/openrouter/example/mimo_example.py