TeleChat-52B: benchmark results

Provider: Other.

Aggregate Rating 1369 ± 28, rank #7193 of 10916 rated models, from 86 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
OpenCompass Reasoning - Deductive - English (CompassBench 2405)27.9Score (%)98.7
OpenCompass Reasoning - Common Sense - English (CompassBench 2405)34.5Score (%)94.7
OpenCompass Language - Content Summarization - English (CompassBench 2405)31.6Score (%)84.2
OpenCompass Language - Intention Recognition - English (CompassBench 2405)79.3Score (%)82.9
OpenCompass LLM - Reasoning - English (CompassBench 2405)42.3Score (%)81.6
OpenCompass LLM - Reasoning (CompassBench 2405)42.6Score (%)76.3
OpenCompass Reasoning - Common Sense - Chinese (CompassBench 2405)39.1Score (%)76.3
OpenCompass Language - Chinese Semantic Understanding - Chinese (CompassBench 2405)61.3Score (%)71.1
OpenCompass Reasoning - Abductive - English (CompassBench 2405)76Score (%)69.7
OpenCompass Language - Intention Recognition - Chinese (CompassBench 2405)58.3Score (%)68.4
OpenCompass Language - Traditional Cultural Understanding - Chinese (CompassBench 2405)57.7Score (%)68.4
OpenCompass LLM - Language - Chinese (CompassBench 2405)52.7Score (%)63.2

Interactive version: theaggregate.ai/model?slug=telechat-52b · How It Works · Data refreshed daily, snapshot 2026-11-08.