HPR-Bench - Tag Inference - Household Context: leaderboard
Metric: Avg@10 Tag Accuracy (%; mean of 10 samples). Source: arxiv.org. Saturation forecast: Around September 2027. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.6 Plus (Thinking) | 59.65 |
| 2 | Qwen 3.5 397B A17B (Thinking) | 54.66 |
| 3 | Qwen 3.5 Plus (Thinking) | 54 |
| 4 | Kimi K2 (Thinking) | 53.8 |
| 5 | Kimi K2.5 (Thinking) | 53.8 |
| 6 | GLM-5 (Thinking) | 53.45 |
| 7 | GLM-4.7 (Thinking) | 52 |
| 8 | DeepSeek V3.2 (Thinking) | 49.81 |
| 9 | DeepSeek R1 0528 | 47.52 |
| 10 | Qwen 3 235B A22B 2507 Instruct | 41.4 |
| 11 | Qwen 3 235B A22B 2507 (Thinking) | 41.35 |
| 12 | Qwen 3 8B (Thinking) | 38.31 |
| 13 | Qwen 3 8B (Non-reasoning) | 32.81 |
Interactive version: theaggregate.ai/benchmark?slug=hpr-bench-tag-inference-household-context · How It Works · Data refreshed daily, snapshot 2026-09-25.