DeepSeek-V2-Lite: benchmark results
Provider: DeepSeek. Access: Open.
Unified ELO 1331 ± 40, rank #2339 of 2656 rated models, from 60 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MemPoison - Poisoned Task | 5.85 | Success Rate (%) | 100 |
| MERA Industrial - ruTXTAgroBench - Crop cultivation technologies | 69.4 | Score (%) | 97.1 |
| MERA Industrial - ruTXTAquaBench - Mariculture | 42.8 | Score (%) | 82 |
| MERA Industrial - ruTXTAgroBench - Crop production | 42 | Score (%) | 80.1 |
| MERA Industrial - ruTXTAgroBench - Farming systems on agricultural landscapes | 38.1 | Score (%) | 68 |
| MERA Industrial - ruTXTAgroBench - Forage production and meadow farming | 31.1 | Score (%) | 65 |
| MERA Industrial - ruTXTAgroBench - Reclamation farming | 35.2 | Score (%) | 63.1 |
| MERA Industrial - ruTXTAquaBench - Ichthyopathology | 30.1 | Score (%) | 62.6 |
| MERA Industrial - ruTXTAquaBench - Feeding of fish and aquatic organisms | 29.7 | Score (%) | 48.5 |
| Open PL LLM - PolQA Open Book (generative, 5-shot) | 79.05 | Levenshtein Similarity (%) | 47.3 |
| Open PL LLM - PolQA Closed Book (generative, 5-shot) | 49.53 | Levenshtein Similarity (%) | 44.5 |
| MERA Industrial - Agriculture | 30.6 | Score (%) | 35.4 |
Interactive version: theaggregate.ai/model?slug=deepseek-v2-lite · How It Works · Data refreshed daily, snapshot 2026-09-19.