Spark-v3.5: benchmark results
Provider: Other.
Aggregate Rating 1539 ± 24, rank #1845 of 10916 rated models, from 65 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Code - HumanEval-X (CompassBench 2404) | 64.3 | Score (%) | 100 |
| OpenCompass Math - Middle - Chinese (CompassBench 2404) | 57.2 | Score (%) | 100 |
| OpenCompass Reasoning - Abductive - Chinese (CompassBench 2404) | 96 | Score (%) | 100 |
| OpenCompass Reasoning - Deductive - Chinese (CompassBench 2404) | 60.5 | Score (%) | 100 |
| OpenCompass Code - MBPP - English (CompassBench 2404) | 77 | Score (%) | 96.7 |
| OpenCompass LLM - Reasoning - Chinese (CompassBench 2404) | 55.6 | Score (%) | 96.7 |
| OpenCompass LLM - Code (CompassBench 2404) | 67.7 | Score (%) | 96 |
| OpenCompass LLM - Code - English (CompassBench 2404) | 70.1 | Score (%) | 96 |
| OpenCompass Subjective - Language (CompassBench 2404) | 56.8 | Score (%) | 96 |
| OpenCompass Code - HumanEval - Chinese (CompassBench 2404) | 66.5 | Score (%) | 95 |
| OpenCompass Code - HumanEval+ - English (CompassBench 2404) | 68.9 | Score (%) | 93.3 |
| OpenCompass Code - MBPP - Chinese (CompassBench 2404) | 61.6 | Score (%) | 93.3 |
Interactive version: theaggregate.ai/model?slug=spark-v3-5 · How It Works · Data refreshed daily, snapshot 2026-11-08.