YiSM-34B-0rn: benchmark results
altomek's self-merge of Yi-1.5-34B with its Chat variant, aiming to follow instructions while preserving the base model's character. Provider: altomek. Released 2024-05-26. Access: Open.
Aggregate Rating 1522 ± 22, rank #709 of 1656 rated models, from 25 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 75.82 | Accuracy (%) (5-shot) | 99.3 |
| Open LLM Leaderboard v1 - MMLU | 78.51 | Accuracy (%) (5-shot) | 99.3 |
| Open Chinese LLM - GSM8K | 70.36 | Accuracy (%) | 96.9 |
| Open LLM Leaderboard - GPQA | 37.16 | Score | 92.9 |
| Open Portuguese LLM - FaQuAD NLI | 81.68 | Macro F1 (%) | 92.7 |
| Open Chinese LLM - ARC Challenge | 62.12 | Accuracy (%) | 91.9 |
| Open Chinese LLM - C-Eval Semantic | 89.81 | Accuracy (%) | 91.9 |
| Open Chinese LLM - WinoGrande | 70.01 | Accuracy (%) | 91.6 |
| Open Chinese LLM - HellaSwag | 69.1 | Accuracy (%) | 90.8 |
| Open Chinese LLM - CMMLU | 71.04 | Accuracy (%) | 90.5 |
| Open Chinese LLM Leaderboard | 68.36 | Official Nonzero-Task Mean (%) | 90.2 |
| Open LLM Leaderboard v1 - WinoGrande | 83.66 | Accuracy (%) (5-shot) | 90.1 |
Interactive version: theaggregate.ai/model?slug=yism-34b-0rn · How It Works · Data refreshed daily, snapshot 2026-10-20.