# The Multivac — Evaluation Report

**Evaluation ID:** EVAL-20260207-133847
**Date:** Feb 18, 2026
**Category:** reasoning
**Question ID:** REASON-006

---

## Question

Schedule a one-day conference with these constraints:

TALKS: A (90min), B (60min), C (45min), D (30min), E (30min), F (45min)
ROOMS: Main Hall (capacity 500), Room 2 (capacity 100), Room 3 (capacity 50)
TIME: 9:00 AM - 5:00 PM, with mandatory lunch break 12:00-1:00 PM

CONSTRAINTS:
1. Talk A must be in Main Hall (expected attendance: 400)
2. Talk B and C cannot overlap (same speaker)
3. Talk D must be before Talk E (E builds on D's content)
4. Talk F requires Room 2's AV equipment
5. No room can have more than 3 talks total
6. At least one talk must be running at all times (except lunch)
7. Talk A cannot start before 10:00 AM (speaker arriving late)
8. Talk E must end by 3:00 PM (speaker leaving early)

Find a valid schedule or prove none exists.

---

## Winner

**GPT-OSS-120B** (OpenAI)
- Winner Score: 8.31
- Matrix Average: 5.78
- Total Judgments: 90

---

## Rankings

| Rank | Model | Provider | Avg Score | Judgments |
|------|-------|----------|-----------|----------|
| 1 | GPT-OSS-120B | OpenAI | 8.31 | 7 |
| 2 | Claude Opus 4.5 | Anthropic | 7.00 | 6 |
| 3 | MiMo-V2-Flash | Xiaomi | 6.81 | 7 |
| 4 | DeepSeek V3.2 | DeepSeek | 6.80 | 6 |
| 5 | Claude Sonnet 4.5 | Anthropic | 6.40 | 6 |
| 6 | Gemini 3 Flash Preview | Google | 6.09 | 7 |
| 7 | Grok 3 (Direct) | xAI | 5.50 | 6 |
| 8 | OLMo Think | Allen AI | 4.55 | 2 |
| 9 | Gemini 2.5 Flash | Google | 4.36 | 8 |
| 10 | Gemini 3 Pro Preview | Google | 1.94 | 7 |

---

## 10×10 Judgment Matrix

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

| Judge ↓ / Resp → | OLMo Think | MiMo-V2-Flash | Gemini 3 | Grok 3 | Claude Sonnet | DeepSeek V3.2 | Claude Opus | Gemini 3 | Gemini 2.5 | GPT-OSS-120B |
|---|---|---|---|---|---|---|---|---|---|---|
| OLMo Think | — | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| MiMo-V2-Flash | 1.6 | — | 3.9 | 4.6 | 6.4 | 7.5 | 5.8 | 1.6 | 5.5 | 2.8 |
| Gemini 3 | 0.0 | 8.2 | — | 7.7 | 4.5 | 8.2 | 8.6 | 4.2 | 5.0 | 10.0 |
| Grok 3 | 0.0 | 8.2 | 8.4 | — | 9.4 | 7.6 | 8.2 | 2.6 | 5.9 | 8.8 |
| Claude Sonnet | 0.0 | 6.0 | 7.2 | 5.7 | — | 7.0 | 8.2 | 0.7 | 4.1 | 9.8 |
| DeepSeek V3.2 | 7.5 | 7.3 | 8.2 | 5.2 | 4.2 | — | 7.6 | 1.6 | 6.2 | 9.5 |
| Claude Opus | 0.0 | 6.0 | 4.8 | 4.3 | 5.4 | 4.0 | — | 1.0 | 3.6 | 7.3 |
| Gemini 3 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | — | 2.0 | 0.0 |
| Gemini 2.5 | 0.0 | 7.8 | 5.2 | 5.6 | 8.6 | 6.5 | 3.6 | 0.0 | — | 10.0 |
| GPT-OSS-120B | 0.0 | 4.3 | 5.0 | 0.0 | 0.0 | 0.0 | 0.0 | 1.9 | 2.6 | — |

---

## 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: REASON-006. 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-20260207-133847/results
Full dataset: https://app.themultivac.com/dashboard/export