benchgap
Other · model

LFM2.5-VL-3B benchmark scores

As of 2026-11-22, LFM2.5-VL-3B (Other) has measured scores on 10 benchmarks and estimated scores on 24 more.

BenchmarkScoreSource
GDPval-AA0.0%estimated ± 5.5 pp, low confidence
IFBench25.8%measured
IFEval82.3%measured
τ²-bench results55.4%estimated ± 12.4 pp, low confidence
AA-MMMU-Pro30.5%estimated ± 1.5 pp, medium confidence
BFCL v432.5%measured
AA-IFBench56.7%estimated ± 7.4 pp, low confidence
MMMU48.4%measured
ParseBench-132.4%estimated ± 8.4 pp, low confidence
ParseBench-219.4%estimated ± 5.0 pp, low confidence
CharXiv61.9%estimated ± 3.3 pp, medium confidence
CharXiv w/o tools69.8%estimated ± 1.2 pp, medium confidence
AI2D_TEST87.6%estimated ± 1.4 pp, medium confidence
CountBench87.3%measured
DynaMath79.2%estimated ± 1.5 pp, low confidence
ERQA39.9%estimated ± 4.2 pp, medium confidence
MathVision34.6%estimated ± 3.0 pp, medium confidence
MLVU (M-Avg)64.5%estimated ± 2.2 pp, medium confidence
MMMU-Pro30.5%measured
OmniDocBench 1.585.5%estimated ± 1.8 pp, low confidence
RealWorldQA73.1%measured
ScreenSpot Pro37.0%estimated ± 4.8 pp, low confidence
SimpleVQA35.4%measured
V*50.2%estimated ± 5.8 pp, low confidence
Video-MME (w/o subtitle)60.2%estimated ± 1.2 pp, medium confidence
Video-MME (with subtitle)53.5%estimated ± 3.0 pp, medium confidence
VideoMMMU43.6%estimated ± 2.3 pp, medium confidence
We-Math61.3%estimated ± 5.5 pp, low confidence
MedXpertQA (MM)65.0%estimated ± 7.1 pp, low confidence
MMVU39.5%estimated ± 2.7 pp, low confidence
CC-OCR80.1%estimated ± 0.7 pp, medium confidence
RefCOCO (avg)87.9%measured
OCRBench V247.5%measured
PinchBench34.6%estimated ± 5.0 pp, low confidence