ENC-Bench - Point Features: leaderboard
Metric: Accuracy (%) on ENC-Bench's 7,830 point features questions (Perception: reading the attributes of point features such as buoys and lights in full chart scenes; 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 March 2027. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 235B A22B Instruct | 51.79 | #264 |
| 2 | Qwen 3 VL 235B A22B (Thinking) | 49.34 | #228 (Qwen 3 VL 235B A22B) |
| 3 | Qwen 3 VL 32B Instruct | 48.77 | #276 |
| 4 | Gemini 2.5 Pro | 45.38 | #145 |
| 5 | GLM-4.5V | 43.44 | #339 |
| 6 | Llama 4 Maverick Instruct | 42.75 | #439 |
| 7 | Gemini 2.5 Flash | 41.89 | #237 |
| 8 | GPT-4o | 36.39 | #333 |
| 9 | Qwen 3 VL 32B (Thinking) | 35.93 | #287 (Qwen 3 VL 32B) |
| 10 | InternVL3-38B | 27.36 | #395 |
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-point-features · How It Works · Data refreshed daily, snapshot 2026-10-11.