Llama-PLLuM-8B-chat: benchmark results
Provider: Meta. Released 2025-02-24. Access: Open.
Unified ELO 1410 ± 1, rank #1097 of 1392 rated models, from 18 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MT-Bench PL - Humanities | 9.5 | Judge Score (0-10) | 67.3 |
| Open LLM Leaderboard - MuSR | 11.86 | Score | 63.9 |
| LLMZSZL Leaderboard | 47.68 | Score | 59.2 |
| MT-Bench PL - Reasoning | 5.35 | Judge Score (0-10) | 49 |
| Polish EQ-Bench | 50.97 | EQ-Bench Score | 45.5 |
| MT-Bench PL - Overall | 6.05 | Judge Score (0-10) | 35.7 |
| Open LLM Leaderboard - IFEval | 35.15 | Score | 33.6 |
| MT-Bench PL - STEM | 7.5 | Judge Score (0-10) | 32.7 |
| MT-Bench PL - Writing | 7.2 | Judge Score (0-10) | 32.7 |
| Open LLM Leaderboard - MMLU-Pro | 19.1 | Score | 29.2 |
| CPTU Bench | 2.92 | Average Score (1-5) | 25.7 |
| Open LLM Leaderboard - BBH | 16.28 | Score | 24.7 |
Interactive version: theaggregate.ai/model?slug=llama-pllum-8b-chat · How It Works · Data refreshed daily, snapshot 2026-09-05.