VG-LLM Release v3 - CVBenchDepth: leaderboard
VGLLM originalrelease v3; QA checkpoint uses Qwen2.5VL7BInstruct plus VGGT geometry encoder, greedy32uniformframes VSI-Bench. Different3Dscenecheckpoint and later S2/S1+S2 trainedablations are distinct; no generic8Balias for them. S1finalcheckpoint row4; source singleton retained. Metric: Accuracy (%). Higher is better. Source: https://arxiv.org/html/2505.24625v3
Metric: Accuracy (%). Source: arxiv.org. 1 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | VG-LLM-8B | 92.3 |
Interactive version: theaggregate.ai/benchmark?slug=vg-llm-release-v3-cvbenchdepth · How It Works · Data refreshed daily, snapshot 2026-10-17.