# The Multivac — Evaluation Report

**Evaluation ID:** EVAL-20260207-144351
**Date:** Feb 12, 2026
**Category:** analysis
**Question ID:** ANALYSIS-005

---

## Question

Your team ran an A/B test on a checkout flow. Here are the results:

Control (A): 10,000 visitors, 320 conversions (3.2%)
Treatment (B): 10,000 visitors, 380 conversions (3.8%)

The product manager says: "B wins! Let's ship it - that's an 18.75% improvement!"

1. Calculate the statistical significance (provide p-value)
2. What's the 95% confidence interval for the true difference?
3. The test ran for 2 days. What concerns does this raise?
4. You discover Treatment B had a bug on iOS that blocked 2,000 users from even reaching checkout. How does this change your analysis?
5. What would you recommend?

---

## Winner

**MiMo-V2-Flash** (Xiaomi)
- Winner Score: 9.69
- Matrix Average: 8.47
- Total Judgments: 90

---

## Rankings

| Rank | Model | Provider | Avg Score | Judgments |
|------|-------|----------|-----------|----------|
| 1 | MiMo-V2-Flash | Xiaomi | 9.69 | 7 |
| 2 | Claude Sonnet 4.5 | Anthropic | 9.61 | 7 |
| 3 | DeepSeek V3.2 | DeepSeek | 9.56 | 7 |
| 4 | Claude Opus 4.5 | Anthropic | 9.53 | 7 |
| 5 | Gemini 3 Flash Preview | Google | 9.43 | 7 |
| 6 | Grok 4.1 Fast | xAI | 9.39 | 6 |
| 7 | GPT-OSS-120B | OpenAI | 8.63 | 7 |
| 8 | Gemini 2.5 Flash | Google | 8.33 | 8 |
| 9 | GPT-OSS-Legal | OpenAI | 7.69 | 8 |
| 10 | Gemini 3 Pro Preview | Google | 2.82 | 9 |

---

## 10×10 Judgment Matrix

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

| Judge ↓ / Resp → | MiMo-V2-Flash | Gemini 3 | GPT-OSS-Legal | Gemini 2.5 | GPT-OSS-120B | DeepSeek V3.2 | Claude Sonnet | Claude Opus | Gemini 3 | Grok 4.1 Fast |
|---|---|---|---|---|---|---|---|---|---|---|
| MiMo-V2-Flash | — | 8.6 | 7.3 | 8.8 | 8.6 | 9.0 | 9.3 | 9.3 | 2.6 | 9.6 |
| Gemini 3 | 10.0 | — | 8.1 | 9.4 | 8.8 | 9.8 | 10.0 | 10.0 | 4.4 | 9.6 |
| GPT-OSS-Legal | 9.0 | 8.8 | — | 7.0 | 0.0 | 8.8 | 8.8 | 8.8 | 2.0 | 0.0 |
| Gemini 2.5 | 10.0 | 10.0 | 9.0 | — | 9.0 | 10.0 | 10.0 | 9.0 | 2.9 | 9.0 |
| GPT-OSS-120B | 0.0 | 0.0 | 5.8 | 6.5 | — | 0.0 | 0.0 | 0.0 | 2.5 | 0.0 |
| DeepSeek V3.2 | 9.6 | 9.6 | 8.6 | 9.3 | 8.8 | — | 9.6 | 10.0 | 1.5 | 9.3 |
| Claude Sonnet | 9.8 | 9.6 | 7.9 | 9.2 | 8.7 | 9.8 | — | 9.6 | 2.1 | 9.6 |
| Claude Opus | 9.6 | 9.6 | 6.7 | 8.0 | 8.1 | 9.6 | 9.6 | — | 1.6 | 9.3 |
| Gemini 3 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | — | 0.0 |
| Grok 4.1 Fast | 10.0 | 10.0 | 8.2 | 8.4 | 8.4 | 10.0 | 10.0 | 10.0 | 5.8 | — |

---

## 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-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-20260207-144351/results
Full dataset: https://app.themultivac.com/dashboard/export