CrashSight - Post-Crash Outcome: leaderboard

Post-crash outcome questions (road user status, vehicle condition, emergency response) of CrashSight (2026), a multiple-choice VQA benchmark on real roadside-camera traffic crash videos answered zero-shot from four frames. Accuracy in percent; higher is better.

Metric: Accuracy (%) on post-crash outcome questions (road-user status, vehicle condition, emergency response), four-option multiple-choice VQA on held-out roadside-camera crash videos of CrashSight-VQA (4 frames at 1 fps, zero-shot, deterministic decoding, exact letter match); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1InternVL3-8B80.3#1782
2Qwen 2.5 VL 7B Instruct74.2#1902
3InternVL3-2B71.2#3290
4Qwen 2.5 VL 3B Instruct62.8#2713
5LLaVA-NeXT-Video-7B57.1#5734
6LLaVA-OneVision-0.5B48#7956

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

Interactive version: theaggregate.ai/benchmark?slug=crashsight-post-crash-outcome · How It Works · Data refreshed daily, snapshot 2026-11-21.