# The Multivac — Evaluation Report

**Evaluation ID:** EVAL-20260402-213909
**Date:** Apr 02, 2026
**Category:** analysis
**Question ID:** ANALYSIS-030

---

## Question

A city's housing data shows: median price $800K (up 40% in 3 years), median income $75K, price-to-income ratio 10.7 (national avg 5.2), rental yield 3.1%, vacancy rate 1.2%, new construction permits down 30% YoY. (1) Is this a bubble? What additional data would you need? (2) Who benefits and who is harmed by these conditions? (3) A politician proposes rent control. Analyze the likely outcomes using economic theory and empirical evidence.

---

## Winner

**Grok 4.20** (openrouter)
- Winner Score: 9.13
- Matrix Average: 8.64
- Total Judgments: 89

---

## Rankings

| Rank | Model | Provider | Avg Score | Judgments |
|------|-------|----------|-----------|----------|
| 1 | Grok 4.20 | openrouter | 9.13 | 9 |
| 2 | GPT-5.4 | openrouter | 9.10 | 9 |
| 3 | MiniMax M2.5 | openrouter | 9.06 | 8 |
| 4 | MiMo-V2-Flash | Xiaomi | 8.96 | 9 |
| 5 | Gemini 3 Flash Preview | Google | 8.87 | 9 |
| 6 | Claude Sonnet 4.6 | openrouter | 8.64 | 9 |
| 7 | DeepSeek V4 | openrouter | 8.59 | 9 |
| 8 | GPT-OSS-120B | OpenAI | 8.35 | 9 |
| 9 | Claude Opus 4.6 | openrouter | 8.31 | 9 |
| 10 | Gemini 3.1 Pro | openrouter | 7.38 | 9 |

---

## 10×10 Judgment Matrix

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

| Judge ↓ / Resp → | MiMo-V2-Flash | Gemini 3.1 Pro | Claude Opus | GPT-5.4 | DeepSeek V4 | Claude Sonnet | Grok 4.20 | GPT-OSS-120B | Gemini 3 | MiniMax M2.5 |
|---|---|---|---|---|---|---|---|---|---|---|
| MiMo-V2-Flash | — | 7.8 | 7.8 | 9.2 | 8.8 | 9.0 | 9.0 | 9.4 | 9.0 | 9.2 |
| Gemini 3.1 Pro | 10.0 | — | 7.7 | 9.3 | 9.8 | 8.3 | 10.0 | 7.2 | 10.0 | 10.0 |
| Claude Opus | 8.9 | 6.8 | — | 9.2 | 8.0 | 8.8 | 9.6 | 8.2 | 8.8 | 9.2 |
| GPT-5.4 | 8.6 | 6.1 | 7.3 | — | 8.0 | 7.0 | 8.8 | 6.9 | 8.6 | 8.8 |
| DeepSeek V4 | 9.0 | 9.0 | 9.0 | 9.0 | — | 9.0 | 8.8 | 9.0 | 9.0 | 9.0 |
| Claude Sonnet | 8.8 | 7.3 | 8.8 | 9.2 | 8.0 | — | 9.4 | 8.8 | 8.6 | 9.0 |
| Grok 4.20 | 8.7 | 8.1 | 8.7 | 9.0 | 8.4 | 8.8 | — | 8.8 | 8.6 | 8.8 |
| GPT-OSS-120B | 8.4 | 6.3 | 7.3 | 8.7 | 8.4 | 8.7 | 8.7 | — | 8.4 | 8.4 |
| Gemini 3 | 9.8 | 8.4 | 9.3 | 9.6 | 9.6 | 9.8 | 9.8 | 9.6 | — | · |
| MiniMax M2.5 | 8.4 | 6.8 | 9.0 | 8.8 | 8.2 | 8.4 | 8.2 | 7.3 | 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: ANALYSIS-030. 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-20260402-213909/results
Full dataset: https://app.themultivac.com/dashboard/export