Spark-v4.0-Ultra: benchmark results
Provider: Other.
Aggregate Rating 1561 ± 28, rank #1527 of 10916 rated models, from 159 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Math - Competition - Chinese (CompassBench 2501) | 25 | Score (%) | 84.6 |
| OpenCompass Math - College - Chinese (CompassBench 2411) | 39.8 | Score (%) | 74.2 |
| OpenCompass Language - Dialogue - Chinese (CompassBench 2501) | 69.3 | Score (%) | 71.8 |
| OpenCompass Math - Competition - English (CompassBench 2501) | 20 | Score (%) | 69.2 |
| OpenCompass Reasoning - Common - Chinese (CompassBench 2501) | 65.2 | Score (%) | 69.2 |
| OpenCompass Knowledge - Engineering - English (CompassBench 2411) | 81.7 | Score (%) | 68.2 |
| OpenCompass LLM - Math - Chinese (CompassBench 2501) | 24.8 | Score (%) | 66.7 |
| OpenCompass Reasoning - Academic - English (CompassBench 2501) | 53.2 | Score (%) | 65.4 |
| OpenCompass Math - Arithmetic (CompassBench 2411) | 82 | Score (%) | 65.2 |
| OpenCompass LLM - Instruction Following - Chinese (CompassBench 2501) | 54.2 | Score (%) | 64.1 |
| OpenCompass LLM - Math (CompassBench 2501) | 25.1 | Score (%) | 64.1 |
| OpenCompass Reasoning - Academic - Chinese (CompassBench 2501) | 61.5 | Score (%) | 64.1 |
Interactive version: theaggregate.ai/model?slug=spark-v4-0-ultra · How It Works · Data refreshed daily, snapshot 2026-11-08.