MMNeedle (1 Image, 4*4 Stitching, Exact Accuracy): leaderboard
MMNeedle multimodal needle-in-a-haystack benchmark (1 Image, 4*4 stitching): exact-match accuracy at locating a target sub-image within stitched image grids.
Metric: 1 Image, 4*4 Stitching, Exact Accuracy. Source: huggingface.co. Saturation forecast: Estimated already saturated. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 83 | #343 |
| 2 | Gemini 1.5 Pro | 39.85 | #445 |
| 3 | Gemini 1.0 Pro | 24.78 | #933 |
| 4 | Claude 3 Opus | 12.3 | #571 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mmneedle-1-image-4-star4-stitching-exact-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-20.