HM-Bench (Report): leaderboard
HM-Bench (2026) tests multimodal LLMs on hyperspectral image understanding with 19,337 multiple-choice questions over 2,178 samples in 13 perception and reasoning tasks, given report-only input: a structured text report of spectral and spatial statistics computed from the cube. Accuracy (%), zero-shot; higher is better.
Metric: Average accuracy (%) over all tasks of HM-Bench's 19,337 hyperspectral multiple-choice questions (2,178 samples, 13 task categories) with report-only input: a structured text report of spectral and spatial statistics computed from the cube; zero-shot, one option letter as output, temperature 0, at most 64 new tokens; higher is better. Source: arxiv.org. Saturation forecast: Around 2033. 16 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL3.5-14B | 39.52 | #1504 |
| 2 | GeoLLaVA-8K | 38.76 | |
| 3 | GLM-4.6V-Flash | 37.72 | #1040 |
| 4 | Kimi-VL-A3B-Instruct | 37.57 | #2253 |
| 5 | GPT-5.4 Mini | 37.45 | #406 |
| 6 | InternVL3-14B | 37.26 | #1214 |
| 7 | InternVL3-9B | 37.18 | #1728 |
| 8 | Llama-3.1-Nemotron-Nano-VL-8B-V1 | 37 | |
| 9 | LLaVA-Next-8B | 36.71 | #2046 |
| 10 | Claude Sonnet 4.6 | 36.57 | #173 |
| 11 | InternVL2-8B | 36.17 | #3227 |
| 12 | Qwen 3 VL 4B | 35.93 | #1151 |
| 13 | GeoChat | 34.06 | #3155 |
| 14 | InternVL2.5-4B | 32.3 | #2549 |
| 15 | Grok 4 | 31.68 | #319 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=hm-bench-report · How It Works · Data refreshed daily, snapshot 2026-11-21.