Yi Large (Preview) — benchmark results

01.AI preview Yi Large model row. Provider: 01.AI. Released 2024-06-17. Access: API.

Unified ELO 1504 ± 49, rank #785 of 1776 rated models, from 11 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
BenchBench87.14Aggregate Score (%)94.1
WildBench55.29WB Score Task-Macro93.5
AlpacaEval 2.051.89LC Win Rate (%)88.3
MixEval56.8Score80.4
HELM NaturalQuestions (Closed)42.77F1 (%)72.2
HELM Lite51.21Mean win rate (self-reported)53.2
HELM (Stanford)47.1Mean Win Rate (%)45.6
ZebraLogic18.9Puzzle Accuracy (%)44.3
HELM WMT 201417.61BLEU-4 (%)34.4
HELM NaturalQuestions (Open)58.64F1 (%)8.9
HELM NarrativeQA37.28F1 (%)3.3

Interactive version: theaggregate.ai/model?slug=yi-large-preview · How the rankings work · Data refreshed daily, snapshot 2026-07-22.