ENC-Bench - Linestring Features: leaderboard
Metric: Accuracy (%) on ENC-Bench's 2,562 linestring features questions (Perception: identifying the type and direction of marked linear features such as tracks and depth contours; four options), on ENC-Bench's 840 NOAA S-57 electronic navigational charts rendered with OpenCPN in day, dusk and night palettes at six scales, zero-shot; higher is better. Source: arxiv.org. Saturation forecast: Around September 2027. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 30.05 | #145 |
| 2 | Qwen 3 VL 235B A22B Instruct | 29.7 | #264 |
| 3 | Qwen 3 VL 32B Instruct | 29.39 | #276 |
| 4 | Gemini 2.5 Flash | 29.08 | #237 |
| 5 | Qwen 3 VL 32B (Thinking) | 28.45 | #287 (Qwen 3 VL 32B) |
| 6 | Qwen 3 VL 235B A22B (Thinking) | 28.18 | #228 (Qwen 3 VL 235B A22B) |
| 7 | GPT-4o | 21.58 | #333 |
| 8 | GLM-4.5V | 20.92 | #339 |
| 9 | InternVL3-38B | 19.87 | #395 |
| 10 | Llama 4 Maverick Instruct | 19.83 | #439 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=enc-bench-linestring-features · How It Works · Data refreshed daily, snapshot 2026-10-11.