Who leads the FB15k-237 benchmark?

NBFNet currently leads FB15k-237 with a score of 0.41 on Mrr.

What is the state-of-the-art score on FB15k-237?

The state-of-the-art result on FB15k-237 is 0.41 (Mrr), achieved by NBFNet as of 2026.

How many models are tracked on FB15k-237?

Codesota tracks 3 models on FB15k-237.

When was the FB15k-237 leaderboard last updated?

The FB15k-237 leaderboard on Codesota includes results through 2026.

Codesota · Benchmark · FB15k-237Home/Leaderboards/Structured Data & Forecasting/Knowledge Graph Completion/FB15k-237

Unknown

FB15k-237.

Name: FB15k-237 Benchmark Results
Creator: Unknown
Published: 2026-01-01
License: https://creativecommons.org/licenses/by/4.0/

Subset of Freebase with 14,541 entities and 237 relations. Inverse-relation leakage removed vs original FB15k. Standard link-prediction / KG completion benchmark, evaluated by filtered MRR and Hits@10.

Paper ↗Leaderboard ↓

§ 01 · Leaderboard

Results by metric.

Only 3 models on this benchmark

Help build the community leaderboard — submit your model results.

Found a wrong score or missing run?

Use row edits to send a sourced correction into moderation.

Add / edit result ↗Report issue ↗

Mrr

Mrr is the reported evaluation metric for FB15k-237. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Mrrverifiedpapervendorcommunityunverified

Rank	Model	Trust	Score	Year	Links	Fix
01	NBFNet NBFNet filtered MRR on FB15k-237 test. seed — verify	paper	0.41	2026	Source ↗	Looks wrong?
02	ULTRA ULTRA zero-shot filtered MRR on FB15k-237. seed — verify	paper	0.37	2026	Source ↗	Looks wrong?
03	RotatE RotatE filtered MRR on FB15k-237 test. seed — verify	paper	0.34	2026	Source ↗	Looks wrong?

§ 04 · Submit a result

Add to the leaderboard.

← Back to Knowledge Graph Completion