DeepSeek Coder V2 Lite Instruct — benchmark results
DeepSeek's 16B MoE coder (2.4B active) with a 128K context, supporting 338 programming languages. Provider: DeepSeek. Released 2024-06-17. Access: Open.
Unified ELO 1438 ± 29, rank #1051 of 1776 rated models, from 30 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| GSM8K | 87.6 | Accuracy (%) | 88.3 |
| CanAiCode | 100 | Junior-v2 Python Pass Rate (%) | 85.6 |
| LiveBench AMPS Hard | 33 | Score | 75 |
| LiveBench Math Comp | 33.33 | Score | 73.6 |
| LiveBench Table Reformat | 52 | Score | 62.5 |
| LiveBench Coding Completion | 23.68 | Score | 59 |
| LiveBench Web Of Lies V2 | 24 | Score | 54.2 |
| LiveBench LCB Generation | 30 | Score | 53.5 |
| LiveBench Table Join | 9 | Score | 50 |
| LiveBench Olympiad | 36.98 | Score | 44.4 |
| LiveBench Typos | 20 | Score | 44.4 |
| BigCodeBench | 36.8 | Pass@1 (%) | 39.3 |
Interactive version: theaggregate.ai/model?slug=deepseek-coder-v2-lite-instruct · How the rankings work · Data refreshed daily, snapshot 2026-07-22.