SRBench - Yelp (Recall@5): leaderboard
SRBench (2026) evaluates LLMs as sequential recommenders that predict the next item a user interacts with from the full interaction history, using a structured prompt and a numeric answer extractor. This board is Recall@5 from 0 to 1 on Yelp business reviews; higher is better.
Metric: Recall@5 (fraction) of the held-out next item among five recommendations on Yelp, LLM sequential recommendation with the full-length interaction history and SRBench's augmented prompt and extractor; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 7 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 3 | 0.0656 | #611 |
| 2 | SRBench DeepSeek-v3 (checkpoint unspecified) | 0.0332 | |
| 3 | GPT-4.1 | 0.0296 | #505 |
| 4 | GLM-3-Turbo | 0.0215 | |
| 5 | Claude Sonnet 4 | 0.0159 | #410 |
| 6 | Llama-2 (SRBench checkpoint unspecified) | 0.0107 | |
| 7 | Llama-3.2 (SRBench checkpoint unspecified) | 0.0087 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=srbench-yelp-recall-5 · How It Works · Data refreshed daily, snapshot 2026-11-21.