LLaVA-Bench

LLaVA-Bench: Evaluates multimodal understanding across image, text, chart, diagram, or cross-modal reasoning tasks.

3rows
llava_bench_overallprimary metric
2026-05-27sampled

Metadata

ID
llava_bench
Category
Intelligence
Release
Unknown

Metrics

LLaVA-Bench-Conv, LLaVA-Bench-Detail, LLaVA-Bench-Complex, LLaVA-Bench-Overall

Latest Results

Rows are parsed from the public LLaVA model zoo table containing LLaVA-Bench-Overall.

Rank Subject LLaVA-Bench-Overall Model Match Provenance Sampled
1 Vicuna-13B-v1.3 / CLIP-L-336px 70.1 — Imported 2026-05-27
2 LLaMA-2-13B-Chat / CLIP-L 67.9 — Imported 2026-05-27
3 LLaMA-2-7B-Chat / CLIP-L 62.8 — Imported 2026-05-27