ENC-Bench - Polygon Features: leaderboard

Metric: Accuracy (%) on ENC-Bench's 4,110 polygon features questions (Perception: interpreting marked area features such as traffic separation zones and restricted areas; 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 February 2027. 10 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Pro39.95#145
2Gemini 2.5 Flash39.44#237
3InternVL3-38B30.29#395
4Qwen 3 VL 235B A22B Instruct29.93#264
5Qwen 3 VL 235B A22B (Thinking)27.83#228 (Qwen 3 VL 235B A22B)
6Qwen 3 VL 32B Instruct26.86#276
7Qwen 3 VL 32B (Thinking)24.67#287 (Qwen 3 VL 32B)
8GPT-4o23.97#333
9GLM-4.5V21.61#339
10Llama 4 Maverick Instruct18.54#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-polygon-features · How It Works · Data refreshed daily, snapshot 2026-10-11.