Skip to main content
CARLA HQ
PAIRWISE BENCHMARKAdaptive Repeated 3-Fold21 Datasets

EXAONE Tabular vs TabFM Benchmark

Empirical head-to-head evaluation of EXAONE Tabular (LG AI Research) versus Google TabFM (Google Research) on identical, precomputed IID or non-IID splits across 21 diverse tabular datasets.

HEAD-TO-HEAD WINS?Direct pairwise dataset matchup wins where the model achieved lower out-of-sample error loss (1 - AUC for binary, Log-Loss for multiclass, RMSE for regression).
EXAONE Tabular4
:
14TabFM
(3 ties) across 21 benchmarks
TOURNAMENT ELO?Global Bradley-Terry Elo rating computed across all evaluation-split matchups with 1,000 bootstrap resamples. Higher is better (baseline 1000).
EXAONE Tabular1021 (±77)
vs
TabFM1111 (±93)
Higher is better • 95% Bootstrap CI
MACRO IMPROVABILITY?Normalized average error regret relative to empirical suite ceilings across all 21 datasets. Lower is better (0.0% is optimal ceiling).
EXAONE Tabular0.85%
vs
TabFM0.85%
Lower is better • Error regret vs ceiling
MEAN SPLIT LATENCY?Average execution latency in seconds per persisted IID or non-IID evaluation split across all benchmarks. Lower is faster.
EXAONE Tabular158.67s
vs
TabFM742.64s
Lower is faster • adaptive split protocol
ARCHITECTURAL BLUEPRINTS

Model Specifications & Design Trade-offs

Comparing transformer architecture, Bayesian priors, sequence context windows, and browser execution runtime.

LG AI Research

EXAONE Tabular

GitHub / Paper ↗

Large-scale tabular foundation model by LG AI Research designed for high-capacity tabular representation learning, handling mixed categorical and continuous variables across tabular domains.

ArchitectureMulti-Task Columnar Autoregressive Tabular Transformer
Prior / PretrainingHybrid pre-training on synthetic datasets and diverse real-world tabular corpuses
Context WindowTokenized columnar context with sequence truncation
Runtime PlatformServer-side only (Requires Python runtime / PyTorch / CUDA)
LicenseOpen Source (Apache 2.0)
Google Research

Google TabFM

GitHub / Paper ↗

Google Research's tabular foundation model architecture focusing on robust columnar tokenization and zero-shot in-context transfer across diverse tabular benchmarks.

ArchitectureColumnar Embedding & Self-Supervised In-Context Transformer
Prior / PretrainingGenerative synthetic distributions and masked column modeling
Context WindowColumnar token budget per batch
Runtime PlatformServer-side only (Requires Python / JAX / PyTorch environment)
LicenseOpen Source (Apache 2.0)
DEEP DIVE REPORTON-DEVICE RUNTIME ANALYSIS

TabICLv2 vs TabPFN-3 vs TabFM vs EXAONE: Which Model Can Run in the Browser?

Evaluating model weight size (MB vs GB), VRAM requirements, Python server infrastructure overhead, licensing terms, and WebGPU client-side execution compatibility.

HEAD-TO-HEAD BREAKDOWN

21-Dataset Metric Evaluation Matrix

Empirical performance on deterministic, application-appropriate IID or non-IID splits: 1 - AUC for binary classification, Log-Loss for multiclass, and RMSE for regression (all metrics: lower is better).

DatasetDomainTaskRows
Splits?Repeats × folds. IID datasets use three folds with size-aware repeats, and the same row indices are used for every model.
Metric?Primary evaluation error loss: 1 - AUC for binary classification, Log-Loss for multiclass, RMSE for regression. All metrics: lower is better.
EXAONE Tabular
Loss?EXAONE Tabular primary out-of-sample error loss (lower is better).
TabFM
Loss?Google TabFM primary out-of-sample error loss (lower is better).
Margin (Δ)?Error loss difference. Negative (green) indicates EXAONE Tabular lead; positive (red) indicates TabFM lead.
Winner?Model achieving lower out-of-sample error loss on this benchmark split.
EXAONE Tabular
Latency?Mean inference runtime in seconds per evaluation split for EXAONE Tabular.
TabFM
Latency?Mean inference runtime in seconds per evaluation split for TabFM.
Abalone ↗Agriculture, Forestry & FishingREG4,1773×3RMSE2.032.02+0.01TabFM13.78s147.89s
Adult ↗Economics & Public PolicyBIN48,8423×31 - AUC0.06800.0677+0.0003TabFM859.66s4421.56s
Airfoil Self Noise ↗UCIREG1,50310×3RMSE1.161.11+0.05TabFM2.34s43.06s
Amazon Employee Access ↗Information Technology & Enterprise SecurityBIN32,7693×31 - AUC0.12320.1273-0.0040EXAONE Tabular203.54s3389.93s
Bank Marketing ↗Finance & BankingBIN45,2113×31 - AUC0.18760.1840+0.0036TabFM749.39s3095.77s
Blood Transfusion Service Center ↗Healthcare & BiomedicineBIN74820×31 - AUC0.24800.2445+0.0035TabFM0.24s10.38s
Breast W ↗Healthcare & Life SciencesBIN69920×31 - AUC0.00540.0055±0.00Tie0.30s10.88s
Car ↗Automotive & Fleet ManagementMULTI1,72810×3Log-Loss0.03460.0108+0.0238TabFM0.56s25.35s
Compas Two Years ↗Legal & Public SafetyBIN5,2783×31 - AUC0.26120.2619-0.0007EXAONE Tabular4.33s107.91s
Credit G ↗Finance & BankingBIN1,00010×31 - AUC0.19640.1938+0.0026TabFM0.69s19.29s
Diabetes ↗Healthcare & BiomedicineBIN76810×31 - AUC0.16380.1623+0.0014TabFM0.30s11.50s
Employee Salaries ↗Human Resources & Workforce AnalyticsREG9,2283×3RMSE7,42710,435-3,008EXAONE Tabular77.76s466.77s
Fitness Club ↗Fitness, Sports & RecreationBIN1,50010×31 - AUC0.17970.1795±0.00Tie0.49s22.13s
House Sales ↗Real Estate & Property ValuationREG21,6133×3RMSE99,52098,887+633TabFM931.00s1839.00s
Houses ↗Real Estate & Urban PlanningREG20,6403×3RMSE39,97137,481+2,490TabFM459.14s1513.45s
Monks Problems 2 ↗Data Science & Artificial IntelligenceBIN60120×31 - AUC0.00000.0000±0.00Tie0.24s9.21s
Phoneme ↗Speech Processing & Acoustic EngineeringBIN5,4043×31 - AUC0.02760.0202+0.0074TabFM2.54s97.68s
Spambase ↗Cybersecurity & IT InfrastructureBIN4,6013×31 - AUC0.00760.0066+0.0010TabFM12.63s154.06s
Telco Customer Churn ↗Telecommunications & Subscription ServicesBIN7,0433×31 - AUC0.14840.1474+0.0009TabFM11.95s171.54s
Titanic ↗Maritime Safety & Actuarial Risk AnalysisBIN1,30910×31 - AUC0.11460.1166-0.0020EXAONE Tabular0.58s22.09s
Vehicle ↗Automotive Engineering & Computer VisionMULTI84610×3Log-Loss0.23930.1986+0.0407TabFM0.53s15.97s
ZERO-SERVER CLIENT-SIDE EXECUTION

Evaluate Foundation Models In Your Spreadsheets

Carla HQ brings TabICLv2 foundation models directly into Google Sheets via WebGPU. Run zero-trust predictions on private data without API keys or cloud pipelines.

FREQUENTLY ASKED QUESTIONS

Frequently Asked Questions: EXAONE Tabular vs TabFM

Common architectural and benchmarking questions comparing EXAONE Tabular and Google TabFM.

How does EXAONE Tabular compare to TabFM on tabular benchmarks?

Across the Carla benchmark suite, EXAONE Tabular and Google TabFM are evaluated head-to-head on identical, precomputed IID or non-IID splits using standardized error losses (1 - AUC for binary classification, Log-Loss for multiclass, and RMSE for regression). IID datasets use adaptive repeated 3-fold evaluation to reduce variance. TabICLv2 enables 100% private in-browser WebGPU execution, whereas server-side models like TabFM require heavy PyTorch infrastructure and cloud GPU backend servers.

Can EXAONE Tabular or TabFM run in the browser without server infrastructure?

Only TabICLv2 (via Carla HQ and nanotabicl ONNX Runtime WASM/WebGPU) is engineered to execute entirely client-side inside web browsers like Google Chrome. EXAONE Tabular requires full Python and PyTorch server infrastructure with high memory allocations.

What are the primary architectural differences between EXAONE Tabular and TabFM?

EXAONE Tabular uses multi-task columnar autoregressive tabular transformer, whereas Google TabFM leverages columnar embedding & self-supervised in-context transformer. TabICLv2 specifically utilizes prefix attention with KV cache reuse for real-time tabular evaluation in spreadsheets.