CHOICE

Remote-sensing vision-language benchmark for perception and reasoning over Earth-observation imagery across multiple task dimensions.

24rows
overall_scoreprimary metric
2026-05-27sampled

Metadata

ID
choice_rs_vlm
Category
Geospatial
Release
Unknown

Metrics

Overall Score, Perception Score, Reasoning Score, ILC, SII, CID, AttR, AssR, CSR

Latest Results

Rows parsed from the CHOICE project page embedded leaderboardData object. Overall score is the mean of the six published L2 capability scores.

Rank Subject Overall Score Model Match Provenance Sampled
1 Qwen2-VL-72B 0.7378 — Imported 2026-05-27
2 InternVL2-40B 0.7263 — Imported 2026-05-27
3 Ovis1.6-Gemma2-9B 0.6998 — Imported 2026-05-27
4 Gemini-1.5-Pro 0.6942 — Imported 2026-05-27
5 Qwen2-VL-7B 0.6922 — Imported 2026-05-27
6 InternVL2-26B 0.6858 — Imported 2026-05-27
7 InternVL2-8B 0.6772 — Imported 2026-05-27
8 GLM-4V-9B 0.6435 — Imported 2026-05-27
9 GPT-4o-2024-11-20 0.6275 GPT-4o
openai-gpt-4o
 		Imported 2026-05-27
10 Molmo-7B-D 0.6215 — Imported 2026-05-27
11 MiniCPM-V-2.5 0.6163 — Imported 2026-05-27
12 GPT-4o-mini 0.6133 GPT-4o-mini
openai-gpt-4o-mini
 		Imported 2026-05-27
13 DeepSeek-VL-7B 0.6102 — Imported 2026-05-27
14 LLaVA-1.6-13B 0.5907 — Imported 2026-05-27
15 Llama3.2-11B 0.5672 — Imported 2026-05-27
16 mPLUG-Owl3-7B 0.5648 — Imported 2026-05-27
17 VHM 0.5623 — Imported 2026-05-27
18 LLaVA-1.6-7B 0.5592 — Imported 2026-05-27
19 Phi3-Vision 0.5293 — Imported 2026-05-27
20 GeoChat 0.5067 — Imported 2026-05-27
21 LHRS-Bot-nova 0.4930 — Imported 2026-05-27
22 RemoteCLIP 0.4868 — Imported 2026-05-27
23 GeoRSCLIP 0.4532 — Imported 2026-05-27
24 LHRS-Bot 0.4160 — Imported 2026-05-27