ResearchTasksConversational AbilityConversational AbilityFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMT-BenchQwen3-32B7.58MT-Bench Score28Mar 24, 2026Alpaca (test)LoRA71.87Alpaca LC Win Rate20Feb 26, 2026Vicuna Eval (test)FAA8.91Vicuna Eval GPT-4 Score20Feb 26, 2026CharacterEvalGPT-4+PCL*3.612Fluency13Feb 26, 2026