Spark X1: benchmark results
Provider: iFlytek. Released 2025-01-15. Access: API.
Aggregate Rating 1604 ± 28, rank #420 of 1635 rated models, from 74 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Math - Competition - Chinese (CompassBench 2504) | 60 | Score (%) | 77.8 |
| OpenCompass LLM - Instruction Following - Chinese (CompassBench 2504) | 66.7 | Score (%) | 74.1 |
| OpenCompass LLM - Math - Chinese (CompassBench 2504) | 53.8 | Score (%) | 74.1 |
| OpenCompass Math - College - Chinese (CompassBench 2504) | 52 | Score (%) | 74.1 |
| OpenCompass Language - Creation - Chinese (CompassBench 2504) | 85.4 | Score (%) | 70.4 |
| OpenCompass Language - Dialogue - Chinese (CompassBench 2504) | 90.7 | Score (%) | 70.4 |
| OpenCompass Reasoning - Puzzle - Chinese (CompassBench 2504) | 90.8 | Score (%) | 70.4 |
| OpenCompass LLM - Language - Chinese (CompassBench 2504) | 82.1 | Score (%) | 68.5 |
| OpenCompass Code - Competition - Chinese (CompassBench 2504) | 59.6 | Score (%) | 66.7 |
| SuperCLUE General (May 2025) - Text Understanding and Creation | 79.54 | Score | 66.7 |
| OpenCompass Math - Reasoning Computation - Chinese (CompassBench 2504) | 49.4 | Score (%) | 64.8 |
| OpenCompass Knowledge - Science - English (CompassBench 2504) | 83 | Score (%) | 63 |
Interactive version: theaggregate.ai/model?slug=spark-x1 · How It Works · Data refreshed daily, snapshot 2026-10-15.