PLLuM-12B-chat: benchmark results
Provider: PLLuM. Released 2025-02-24. Access: Open.
Aggregate Rating 1414 ± 18, rank #5461 of 10916 rated models, from 61 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLMZSZL Leaderboard | 49.69 | Score | 62.2 |
| MT-Bench PL - Writing | 8 | Judge Score (0-10) | 55.1 |
| MT-Bench PL - Humanities | 9.3 | Judge Score (0-10) | 52 |
| Open LLM Leaderboard - MuSR | 41.15 | Score | 51.7 |
| Belebele European - English | 78.78 | Accuracy (%, 0-shot) | 50 |
| Polish EQ-Bench | 52.26 | EQ-Bench Score (parseability-adjusted) | 49.5 |
| Belebele European - Lithuanian | 62.89 | Accuracy (%, 0-shot) | 46.4 |
| Belebele European - Polish | 68.67 | Accuracy (%, 0-shot) | 46.4 |
| Belebele European - Albanian | 41.33 | Accuracy (%, 0-shot) | 42.9 |
| Belebele European - Ukrainian | 67.89 | Accuracy (%, 0-shot) | 42.9 |
| Belebele European - Russian | 69.78 | Accuracy (%, 0-shot) | 39.3 |
| CPTU Bench | 3.14 | Average Score (1-5) | 38.6 |
Interactive version: theaggregate.ai/model?slug=pllum-12b-chat · How It Works · Data refreshed daily, snapshot 2026-11-08.