T-lite-instruct-0.1 — benchmark results

Provider: Other. Access: Open.

Unified ELO 1358 ± 37, rank #1371 of 1756 rated models, from 14 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LIBRA - LibrusecMHQA48.44Dataset Total Score (%)92.9
Shlepa - Movies MC (RU)46.99Accuracy (%)82.4
Shlepa - Books MC (RU)33.61Accuracy (%)76.5
LIBRA - ruTREC30.72Dataset Total Score (%)64.3
Ru Arena Hard64.66Win Rate (%)63.7
Shlepa - Music MC (RU)28.09Accuracy (%)58.8
Shlepa - Law MC (RU)53.24Accuracy (%)55.9
LIBRA - LongContextMultiQ5.17Dataset Total Score (%)50
LIBRA - ruSciPassageCount5.08Dataset Total Score (%)50
LIBRA - MatreshkaNames14Dataset Total Score (%)42.9
LIBRA - ruBABILongQA38.23Dataset Total Score (%)35.7
LIBRA - ruQasper2.73Dataset Total Score (%)35.7

Interactive version: aibenchmarks.dev/model?slug=t-lite-instruct-0-1 · How the rankings work · Data refreshed daily, snapshot 2026-07-18.