MMCL-Bench - Empirical Discovery and Induction: leaderboard

MMCL-Bench (2026) tests multimodal context learning: a model must learn a task-local rule, procedure or pattern from diagrams, screenshots, manuals or videos and apply it to a new visual instance, graded against expert rubrics. This board is the strict pass rate on the 21 tasks that require inducing a hidden pattern from examples; percent, higher is better.

Metric: Strict task pass rate (%) on the 21 empirical discovery tasks (a hidden pattern must be induced from examples; a task passes only when every rubric item is satisfied), on the 102 MMCL-Bench tasks (49 rule-system, 32 procedural, 21 empirical), in which a model learns a task-local rule, procedure or pattern from visual or mixed-modality teaching context and applies it to a new visual instance, scored against expert rubrics by a rubric-only LLM judge, models at their recommended or default decoding settings; higher is better. Source: arxiv.org. Saturation forecast: Around September 2027. 5 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-5.4Thinking23.8#135
2Gemini 3.1 Pro (Preview)19#99
3Claude Opus 4.6Thinking9.5#113
4Qwen 3.6 PlusThinking4.8#238
5Kimi K2.5Thinking0#266

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=mmcl-bench-empirical-discovery-and-induction · How It Works · Data refreshed daily, snapshot 2026-11-21.