MemoryCD (Home & Kitchen) - Review Generation: leaderboard

Metric: ROUGE-L (0-1, times 100) of the generated full review against the user's own review text, on MemoryCD's Home & Kitchen domain (110 users with long histories, their 3 most recent interactions each), the whole history concatenated into the prompt; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 7 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Pro17.8#145
2Gemini 2.5 Flash16.3#237
3Claude Sonnet 414.5#194
4GPT-5 Mini14.4#176
5Claude Opus 414.4#155
6GPT-513.4#91
7GPT-5 Nano10.7#415

No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.

Interactive version: theaggregate.ai/benchmark?slug=memorycd-home-kitchen-review-generation · How It Works · Data refreshed daily, snapshot 2026-10-11.