IPIBench - Proactive Task Management: leaderboard

IPIBench (2026) streams videos to multimodal LLMs at one frame per second and checks whether they respond proactively at the right moment, manage monitoring tasks the user cancels or changes, and handle reactive questions interleaved with proactive tasks. This board is accuracy on the proactive task management tasks (mean of cancellation, modification and multi-task management); percent, higher is better.

Metric: Accuracy (%) on the proactive task management tasks (mean of cancellation, modification and multi-task management), on IPIBench (1,831 videos, 3,738 interaction instances from egocentric and exocentric datasets) under a unified streaming protocol at 1 FPS (offline models see a sliding window of the latest 16 frames); a proactive response counts only if the first trigger falls within one second of the ground-truth moment, multi-turn tasks only if every required trigger and answer is correct, reactive answers by bidirectional substring matching against up to 20 reference answers; higher is better. Source: arxiv.org. Saturation forecast: Around April 2028. 15 models tracked. Updated 14 Nov 2026.

Top models

#ModelScoreOverall rank
1GPT-5.423.5#135
2Gemini 2.5 Pro20.71#276
3Gemini 3 Pro20.22#145
4GPT-4o19.21#751
5Qwen 3.5 Plus18.46#242
6GLM-4.6V16.38#710
7MOSS-VL-Realtime11.46
8Qwen 3.5 9B (Harness: AURA2 IPIBench offline30frames; Thinking: On)11.31
9Qwen 3 VL 8B9.35#917
10JoyAI-VL-Interaction5.99
11InternVL3-8B5.51#1782
12VideoLLM-online-8B5.39
13Qwen 3.5 9B (Harness: AURA2 IPIBench offline30frames; Thinking: Off)3.32
14LLaVA-OneVision-7B0.96#2981
15Flash-VStream-7B0.77

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

Interactive version: theaggregate.ai/benchmark?slug=ipibench-proactive-task-management · How It Works · Data refreshed daily, snapshot 2026-11-21.