Starling-RM-34B: benchmark results
Provider: Other. Access: Open.
Unified ELO 1503 ± 47, rank #731 of 1605 rated models, from 17 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| RewardBench Prior Sets (0.5 weight) | 71.37 | Score (%) | 85.6 |
| RewardBench Chat | 96.93 | Accuracy (%) | 83 |
| RewardBench | 81.33 | Score (%) | 77.5 |
| RewardBench Safety | 87.7 | Accuracy (%) | 74.8 |
| RewardBench Reasoning | 88.45 | Accuracy (%) | 66.9 |
| Themis-CodeRewardBench - Security Hardness | 68.72 | Preference accuracy (%; share of the benchmark's preference | 63.3 |
| Themis-CodeRewardBench - Memory Efficiency | 63.67 | Preference accuracy (%; share of the benchmark's preference | 50 |
| RewardBench Math | 61.75 | Score (%) | 49 |
| RewardBench Chat Hard | 57.24 | Accuracy (%) | 38.4 |
| Themis-CodeRewardBench - Readability and Maintainability | 64.58 | Preference accuracy (%; share of the benchmark's preference | 38 |
| Themis-CodeRewardBench | 70.63 | Preference accuracy (%; share of the benchmark's preference | 32 |
| Themis-CodeRewardBench - Functional Correctness | 77.75 | Preference accuracy (%; share of the benchmark's preference | 29.6 |
Interactive version: theaggregate.ai/model?slug=starling-rm-34b · How It Works · Data refreshed daily, snapshot 2026-09-26.