HAVEN - Joint Summary Keyframes: leaderboard

Joint summarization on HAVEN (2026), a video understanding benchmark with aligned frames, shots, descriptions and full video documents: the model writes a text summary and selects key frames together, and this board scores the frames by F1 against human-annotated keyframes in percent; higher is better.

Metric: Keyframe selection F1 (%) against human-annotated keyframes when summarizing a video jointly into text and frames (video-to-video-and-text), 15 percent frame budget, zero-shot on a random sample of 100 videos per source dataset; higher is better. Source: arxiv.org. Saturation forecast: Around March 2027. 4 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1VideoLLaMA3-7B26.7#2064
2Qwen 3 VL 8B Instruct23.5#1043
3Qwen 2.5 VL 7B Instruct17.6#1919
4InternVideo2.5-Chat-8B6.5#2049

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=haven-joint-summary-keyframes · How It Works · Data refreshed daily, snapshot 2026-11-22.