Spark X1.5: benchmark results
iFlytek's Spark X1.5 (November 6, 2025), a Mixture-of-Experts deep-reasoning model in the Spark X1 line that supports over 130 languages. Provider: iFlytek. Released 2025-11-06. Access: API.
Aggregate Rating 1615 ± 35, rank #397 of 1635 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SuperCLUE-Mkt - Compliance Assessment | 90.83 | Score | 90.9 |
| SuperCLUE-Mkt - Market Insight | 84.39 | Score | 81.8 |
| SuperCLUE-Mkt - Overall | 83.42 | Score | 81.8 |
| SuperCLUE-Mkt - Placement Strategy | 85.73 | Score | 81.8 |
| SuperCLUE-Mkt - Creative Production | 72.72 | Score | 63.6 |
| SuperCLUE General (2025 Annual) - Science Reasoning | 63.11 | Score | 34.8 |
| SuperCLUE General (2025 Annual) - Math Reasoning | 67.83 | Score | 33.3 |
| LLM2014 Logic 2025-11 | 30.7 | Median Score | 32.7 |
| SuperCLUE General (2025 Annual) - Agentic Task Planning | 47.83 | Score | 27.3 |
| SuperCLUE General (2025 Annual) - Hallucination Control | 62.26 | Score | 27.3 |
| LLM2014 Logic 2025-12 | 25.98 | Median Score | 24 |
| LLM2014 Logic 2026-01 | 25.9 | Median Score | 22.7 |
Interactive version: theaggregate.ai/model?slug=spark-x1-5 · How It Works · Data refreshed daily, snapshot 2026-11-05.