← All models

Model profile

LightPFN 1.5.1

LightPFN is a compact transformer pretrained only on synthetic tables, with separate classification and regression checkpoints under five million parameters. It uses labelled training rows as context without updating its weights. The benchmark uses four-member classification and two-member regression ensembles. Classification supports up to 10 classes.

Tabular FoundationRegular feature limit: 500Source ↗

Model profile

LightPFN 1.5.1

Default parameters. Peer models use their prespecified benchmark-default configuration without per-dataset tuning. This card describes that configuration, not the best achievable tuned model.

Operating map

Macro-F1 Elo relative to Random Forest. Columns are training-sample caps; rows are dataset modalities. Small labels show the number of binding targets.

Elo difference vs. Random Forest
Fixed nonlinear scale; extremes capped at ±1,000 Elo.
No result / comparison unavailable

Reliability

Failure rate map

Failed fits / (successful + failed fits) for the selected model and modality. Design skips are omitted. A dash means no attempted fits; 0% means all attempted fits succeeded. Hover for counts. Colours span 0–20%, with rates above 20% using the darkest shade; cell labels and hover retain the actual rate. This chart is independent of the performance sensitivity view.

Reference benchmark results

Strict Macro-F1 Elo at p=10,000, n=100. Random Forest is anchored at 1,000; intervals are target-bootstrap 95% intervals. These reference results stay fixed while you explore other budgets in the charts. Overlapping intervals do not establish a clear winner.

Dataset modalityElo95% intervalTargets
All modalities1,1031,048–1,16143
DNA methylation1,062984–1,1744
GLM/PLM Embeddings1,271823–1,8024
Gene expression1,076975–1,1729
Genomic prediction907578–1,1024
Metagenomics1,0861,013–1,16410
Molecular properties1,1971,120–1,28711
Other1,0911,091–1,0911