HM-Bench (VSR2): 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 full VSR2 input: the RGB, PCA and report views together with the authors' fixed cross-view fusion prompt. 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 full VSR2 input: the RGB, PCA and report views together with the authors' fixed cross-view fusion prompt; zero-shot, one option letter as output, temperature 0, at most 64 new tokens; higher is better. Source: arxiv.org. Saturation forecast: Around 2032. 17 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL3.5-14B | 45.09 | #1519 |
| 2 | GPT-5.4 Mini | 44.28 | #409 |
| 3 | Claude Sonnet 4.6 | 44.01 | #179 |
| 4 | GLM-4.6V-Flash | 43.61 | #1049 |
| 5 | Qwen 3 VL 4B | 43.23 | #1163 |
| 6 | GeoLLaVA-8K | 41.94 | |
| 7 | LLaVA-v1.5-7B | 40.6 | #7642 |
| 8 | InternVL3-9B | 40.47 | #1746 |
| 9 | Kimi-VL-A3B-Instruct | 40.36 | #2272 |
| 10 | InternVL3-14B | 40.05 | #1227 |
| 11 | Llama-3.1-Nemotron-Nano-VL-8B-V1 | 39.61 | |
| 12 | GeoChat | 39.6 | #3175 |
| 13 | Grok 4 | 39.5 | #321 |
| 14 | Qwen 2.5 VL 7B Instruct | 39.41 | #1919 |
| 15 | LLaVA-Next-8B | 39.28 | #2058 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=hm-bench-vsr2 · How It Works · Data refreshed daily, snapshot 2026-11-22.