TriggerBench - Positive Clean PM Accuracy: leaderboard
TriggerBench (2026) tests prospective memory in conversation: a user states a constraint early in a dialogue, and a later turn creates a situation where a good assistant should act on it unprompted; a judge model scores whether each reply does. Dialogues are about 2.5K tokens long. The board is judged accuracy in percent on tasks whose trigger calls for the reminder; higher is better.
Metric: PM accuracy (%) on the 488 positive-clean tasks, where the trigger requires the reminder; base dialogues of about 2.5K tokens; TriggerBench: 1,265 prospective-memory tasks from 488 dialogue blueprints in five dimensions (state tracking, temporal grounding, logical adherence, attention recovery, safe coding); a constraint is stated early in a dialogue and a later trigger turn calls for acting on it without being asked; GPT-4o (T=0) judges whether the reply fulfils the proactive intent; temperature 0.6; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 (High) | 98.16 | #176 (GPT-5.2) |
| 2 | GPT-5.2 (Medium) | 96.72 | #176 (GPT-5.2) |
| 3 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 92.83 | #434 |
| 4 | GPT-5.2 (Non-reasoning) | 89.75 | #176 (GPT-5.2) |
| 5 | Gemma 3 27B (IT) | 76.43 | #1289 |
| 6 | GPT-4.1 | 73.36 | #495 |
| 7 | Qwen 3 235B A22B 2507 Instruct FP8 | 68.65 | #379 |
| 8 | GPT-4o | 62.91 | #741 |
| 9 | Qwen 3 32B | 49.8 | #960 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=triggerbench-positive-clean-pm-accuracy · How It Works · Data refreshed daily, snapshot 2026-11-13.