SenseNova V6 Reasoner: benchmark results

Provider: SenseTime. Released 2025-04-10. Access: API.

Unified ELO 1711 ± 40, rank #309 of 2656 rated models, from 21 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
SuperCLUE General (May 2025) - Agent69.59Score92.3
SuperCLUE General (July 2025) - Agent83.58Score90.2
SuperCLUE General (May 2025) - Math Reasoning62.1Score89.7
SuperCLUE General (May 2025) - Overall62.96Score87.2
SuperCLUE General (May 2025) - Text Understanding and Creation80.43Score87.2
SuperCLUE General (July 2025) - Code Generation80.2Score80.4
SuperCLUE General (May 2025) - Precise Instruction Following31.93Score79.5
SuperCLUE General (May 2025) - Science Reasoning47.52Score73.1
LLM2014 Logic 2025-0455.28Median Score71.4
SuperCLUE General (July 2025) - Overall60.73Score69.6
MathVision55.39Overall Accuracy (%)69.4
SuperCLUE General (May 2025) - Code Generation86.19Score69.2

Interactive version: theaggregate.ai/model?slug=sensenova-v6-reasoner · How It Works · Data refreshed daily, snapshot 2026-09-19.