MSQA - Spanish: leaderboard
MSQA asks short culture-specific questions written by native speakers in their own language, each with one verifiable answer grounded in local sources. This board covers the Spanish-language questions. Answers are free-form, judged for containing the gold answer, and scored as the harmonic mean of the correct share and the correct-given-attempted share, so both wrong answers and refusals cost points (ByteDance Seed, Nanjing University and M-A-P, 2026).
Metric: F-score (%; harmonic mean of CO, the share of fully correct answers, and CGA, the correct share of attempted answers, over the 80 Spanish-language questions; free-form answers to natively sourced culture-specific short questions asked in their own language, five runs per model, scored by a Gemini-3.1-Pro gold-answer containment judge). Source: arxiv.org. Saturation forecast: Around February 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 68.2 | #94 |
| 2 | Claude Opus 4.6 | 63.9 | #102 |
| 3 | Claude Opus 4.7 | 61.6 | #79 |
| 4 | DeepSeek V4 Pro | 56.1 | #188 |
| 5 | Seed 2.1 Pro | 55.1 | #135 |
| 6 | GPT-5.5 | 54.8 | #49 |
| 7 | Kimi K2.5 | 54.3 | #256 |
| 8 | GPT-5.4 | 52.2 | #123 |
| 9 | Kimi K2.6 | 52.1 | #178 |
| 10 | Seed 2.0 Pro (High) | 52.1 | #211 (Seed 2.0 Pro) |
| 11 | GLM-5 | 51.8 | #260 |
| 12 | GPT-5.2 (High) | 51.3 | #176 (GPT-5.2) |
| 13 | Qwen 3.5 Plus (Thinking) | 51.2 | #232 (Qwen 3.5 Plus) |
| 14 | Seed 2.0 Pro (Medium) | 50.8 | #211 (Seed 2.0 Pro) |
| 15 | DeepSeek V3.2 | 50.7 | #406 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=msqa-spanish · How It Works · Data refreshed daily, snapshot 2026-11-13.