SDABench - Descriptive (MCQ): leaderboard

Metric: Accuracy (%) of four-option multiple-choice answers with perturbation-based distractors on the held-out SDA-Synth test split (the 400 descriptive analysis instances; zero-shot; 25% chance); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 13 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)94.25
2Claude Sonnet 4.693.75
3GPT-5.493.5
4Kimi K2.592.25
5Qwen 3.5 397B A17B92
6GLM-591.5
7DeepSeek V3.290
8GPT-5.4 Mini85.75
9Qwen 3 235B A22B 2507 Instruct85.5
10DeepSeek R184.25
11Llama 3.3 70B Instruct28
12Llama 3.1 8B Instruct21.5

Interactive version: theaggregate.ai/benchmark?slug=sdabench-descriptive-mcq · How It Works · Data refreshed daily, snapshot 2026-09-29.