WeightBench - Macro SWB (42 tasks): leaderboard

Default difficulty, seed0. Claude2samples/task; DeepSeek/GPT1sample/task. Quarantined/invalid submissions score0 (unreviewed lower bounds). Task scores average across runs then tasks; macro averages5tracks. GPT excludes4 loop Fix tasks; use common38task macro, never pool GPT9taskFix with13taskFix. Metric: Native S_WB (0–1). Source: https://arxiv.org/html/2611.18553v1

Metric: Native S_WB (0–1). Source: arxiv.org. Saturation forecast: Around September 2027. 4 models tracked. Updated 17 Nov 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 5.50.535#7
2Claude Sonnet 5.50.466#21
3Claude Haiku 5.50.402#76
4DeepSeek V4 Flash0.215#265

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=weightbench-macro-swb-42-tasks · How It Works · Data refreshed daily, snapshot 2026-11-17.