讯飞星火V4.0: benchmark results

Provider: Other.

Aggregate Rating 1512 ± 34, rank #2164 of 10916 rated models, from 26 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
SuperCLUE General (August 2024) - Hard57.86Score56.1
SuperCLUE General (March 2025) - Math Reasoning39.82Score54.3
SuperCLUE General (June 2024) - Science71Score51.6
SuperCLUE General (August 2024) - Science70.32Score46.3
SuperCLUE General (August 2024) - Overall66.06Score43.9
SuperCLUE General (December 2024) - Science62.3Score43.2
SuperCLUE General (September 2023) - Open-Ended Questions (OPEN)40.64Score42.1
SuperCLUE General (September 2023) - Overall50.06Score42.1
SuperCLUE General (March 2025) - Code Generation51.22Score40
SuperCLUE General (June 2024) - Hard60Score39.1
SuperCLUE General (October 2024) - Liberal Arts71.55Score39
SuperCLUE General (June 2024) - Overall67Score37.5

Interactive version: theaggregate.ai/model?slug=v4-0 · How It Works · Data refreshed daily, snapshot 2026-11-08.