# The Multivac — Evaluation Report

**Evaluation ID:** EVAL-20260402-122714
**Date:** Feb 10, 2026
**Category:** code
**Question ID:** CODE-005

---

## Question

Convert this Python code to idiomatic Rust. The code must compile, handle errors properly, and follow Rust best practices.

```python
from dataclasses import dataclass
from typing import Optional, List
from datetime import datetime

@dataclass
class Task:
    id: int
    title: str
    completed: bool
    due_date: Optional[datetime]
    tags: List[str]

class TaskManager:
    def __init__(self):
        self.tasks = []
        self.next_id = 1
    
    def add_task(self, title: str, due_date: Optional[datetime] = None, tags: List[str] = None) -> Task:
        task = Task(
            id=self.next_id,
            title=title,
            completed=False,
            due_date=due_date,
            tags=tags or []
        )
        self.tasks.append(task)
        self.next_id += 1
        return task
    
    def complete_task(self, task_id: int) -> bool:
        for task in self.tasks:
            if task.id == task_id:
                task.completed = True
                return True
        return False
    
    def get_overdue(self) -> List[Task]:
        now = datetime.now()
        return [t for t in self.tasks if t.due_date and t.due_date < now and not t.completed]
```

---

## Winner

**Claude Opus 4.6** (openrouter)
- Winner Score: 8.94
- Matrix Average: 7.78
- Total Judgments: 77

---

## Rankings

| Rank | Model | Provider | Avg Score | Judgments |
|------|-------|----------|-----------|----------|
| 1 | Claude Opus 4.6 | openrouter | 8.94 | 8 |
| 2 | GPT-5.4 | openrouter | 8.88 | 8 |
| 3 | Claude Sonnet 4.6 | openrouter | 8.66 | 7 |
| 4 | GPT-OSS-120B | OpenAI | 8.51 | 6 |
| 5 | Grok 4.20 | openrouter | 8.38 | 8 |
| 6 | Gemini 3 Flash Preview | Google | 8.26 | 8 |
| 7 | DeepSeek V4 | openrouter | 7.75 | 9 |
| 8 | MiMo-V2-Flash | Xiaomi | 7.47 | 9 |
| 9 | MiniMax M2.5 | openrouter | 7.42 | 7 |
| 10 | Gemini 3.1 Pro | openrouter | 3.51 | 7 |

---

## 10×10 Judgment Matrix

Rows = Judge, Columns = Respondent. Self-judgments excluded (—).

| Judge ↓ / Resp → | GPT-5.4 | Claude Opus | Gemini 3.1 Pro | Claude Sonnet | Grok 4.20 | DeepSeek V4 | GPT-OSS-120B | Gemini 3 | MiniMax M2.5 | MiMo-V2-Flash |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-5.4 | — | 7.5 | 0.5 | 6.5 | 7.0 | 6.3 | 4.8 | 6.2 | 4.2 | 5.2 |
| Claude Opus | 9.2 | — | 1.0 | 9.2 | 8.3 | 7.5 | 9.2 | 9.0 | 8.0 | 7.0 |
| Gemini 3.1 Pro | 9.8 | 9.4 | — | 9.4 | 8.3 | 9.0 | 8.8 | 9.2 | 6.5 | 6.2 |
| Claude Sonnet | 8.8 | 9.3 | 1.0 | — | 8.8 | 8.0 | 8.8 | 8.8 | 9.2 | 8.0 |
| Grok 4.20 | 8.6 | 8.4 | 2.5 | · | — | 6.8 | · | 7.8 | 7.5 | 7.0 |
| DeepSeek V4 | · | 9.6 | · | 9.6 | 8.6 | — | 9.6 | 8.8 | 9.6 | 8.8 |
| GPT-OSS-120B | 8.8 | 8.8 | · | 8.6 | · | 7.8 | — | · | 6.9 | 7.8 |
| Gemini 3 | 9.8 | 9.8 | 9.8 | · | 9.8 | 9.6 | 9.8 | — | · | 9.8 |
| MiniMax M2.5 | 7.4 | · | 1.2 | 8.4 | 7.5 | 6.5 | · | 7.5 | — | 7.6 |
| MiMo-V2-Flash | 8.6 | 8.6 | 8.6 | 9.0 | 8.6 | 8.2 | · | 8.8 | · | — |

---

## Methodology

- **10×10 Blind Peer Matrix:** All models answer the same question, then all models judge all responses.
- **5 Criteria:** Correctness, completeness, clarity, depth, usefulness (each scored 1–10).
- **Self-judgments excluded:** Models do not judge their own responses.
- **Weighted Score:** Composite of all 5 criteria.

---

## Citation

The Multivac (2026). Blind Peer Evaluation: CODE-005. app.themultivac.com

## License

Open data. Free to use, share, and build upon. Please cite The Multivac when using this data.

Download raw JSON: https://app.themultivac.com/api/evaluations/EVAL-20260402-122714/results
Full dataset: https://app.themultivac.com/dashboard/export