Imaging-101 - End-to-End: leaderboard
Metric: End-to-end success rate (%): share of the 57 expert-verified computational imaging tasks from six scientific domains whose full reconstruction pipeline, written and executed by the agent, meets the task's NCC and NRMSE image-quality thresholds; higher is better. Source: arxiv.org. Saturation forecast: Around September 2028. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 29.8 |
| 2 | GPT-5.4 | 19.3 |
| 3 | Gemini 3.1 Pro (Preview) | 14 |
| 4 | GLM-5 | 12.3 |
| 5 | DeepSeek V3 | 10.5 |
| 6 | Kimi K2.5 | 7 |
| 7 | Qwen 3.6 Plus | 7 |
Interactive version: theaggregate.ai/benchmark?slug=imaging-101-end-to-end · How It Works · Data refreshed daily, snapshot 2026-09-29.