Yi-Coder-9B: benchmark results
01.AI's Yi-Coder-9B, a code-specialized model supporting 52 programming languages with a context length up to 128K tokens, part of the sub-10B-parameter Yi-Coder series. Provider: 01.AI. Released 2024-09-05. Access: Open.
Aggregate Rating 1331 ± 16, rank #1428 of 1656 rated models, from 48 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Beyond Pass Rate | 23.6 | Correctness (%) | 100 |
| MERA - ruHumanEval | 29.33 | pass@1 (%) | 73.7 |
| ShaderMatch | 14.72 | Clone Match Rate (%) | 71.4 |
| MERA - BPS | 97.7 | Accuracy (%) | 69.5 |
| MERA - ruCodeEval | 25.85 | pass@1 (%) | 69.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.75 | Normalized accuracy (%) | 69.5 |
| MERA - LCS | 14.6 | Accuracy (%) | 59.3 |
| MERA - RWSD | 55 | Accuracy (%) | 43.3 |
| Open Portuguese LLM - ASSIN2 RTE | 84.79 | Macro F1 (%) | 35.1 |
| Open Portuguese LLM - OAB Exams | 37.04 | Accuracy (%) | 34.9 |
| Open Korean LLM - Ko-IFEval | 25.23 | Strict accuracy, prompt/instruction mean (%) | 33.8 |
| Open Portuguese LLM - FaQuAD NLI | 52.12 | Macro F1 (%) | 33.7 |
Interactive version: theaggregate.ai/model?slug=yi-coder-9b · How It Works · Data refreshed daily, snapshot 2026-10-20.