Ling-flash-2.0: benchmark results
InclusionAI's speed-oriented open MoE (100B total/6.1B active), matching ~40B dense models at higher speed (September 2025). Provider: InclusionAI. Released 2025-09-17. Access: Open.
Unified ELO 1551 ± 1, rank #429 of 1392 rated models, from 93 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Medmarks - MedHallu Easy | 76.47 | Score (%) | 95.7 |
| Medmarks - MedHallu Hard | 54.44 | Score (%) | 95.7 |
| Medmarks - MedHallu Medium | 72.2 | Score (%) | 95.7 |
| SciEval - Astronomy | 48.12 | Astronomy (%) | 94.4 |
| Medmarks - PubHealthBench Reviewed | 86.49 | Score (%) | 80 |
| Medmarks - PubMedQA | 76.47 | Score (%) | 72.9 |
| Medmarks - Med-HALT Reasoning NOTA | 61.51 | Score (%) | 71.4 |
| AA Omniscience - Software Engineering (SWE) - PHP | 28.57 | Accuracy (%) | 71 |
| SciEval - Earth Sciences | 69.55 | Earth Sciences (%) | 66.7 |
| Medmarks - SuperGPQA Medicine Easy | 54.68 | Score (%) | 65.7 |
| AA Omniscience - Software Engineering (SWE) - Python | 21.76 | Accuracy (%) | 62.1 |
| Medmarks - HEAD-QA v2 | 83.71 | Score (%) | 60 |
Interactive version: theaggregate.ai/model?slug=ling-flash-2-0 · How It Works · Data refreshed daily, snapshot 2026-09-05.