Gemma 2 27B benchmark results
7 results on 7 benchmarks from 4 sources.
Open weights Released: Jun 24, 2024 Updated Oct 22, 2026
Also listed as: gemma-2-27b-it · Gemma 2 Instruct (27B) · Gemma-2-27B-it · google/gemma-2-27b-it · google/gemma-2-27b
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1289 | 291 | Overall, style control on; 75754 votes; 95% CI 1286–1293“gemma-2-27b-it” | Official leaderboard | Oct 8, 2026 |
| Open LLM Leaderboard v2 | 36.2% | 555 | Average of 6 normalised benchmarks; bfloat16; 27.227B params“google/gemma-2-27b-it” | Official leaderboard | Mar 13, 2025 |
| GPQA Diamond | 36.5% | 245 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
| MMLU | 75.7% | 38 | Exact match; 5-shot | Official leaderboard via Epoch AI | Jun 24, 2024 |
| MMLU-Pro | 56.5% | 149 | Overall accuracy; data source: TIGER-Lab; 27B“Gemma-2-27B-it” | Official leaderboard | Mar 11, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 1.4% | 254 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Mar 7, 2025 |
| MATH Level 5 | 27.9% | 77 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
FAQ
Where do these Gemma 2 27B scores come from?
From Epoch AI Benchmarking Hub, LMArena leaderboard dataset, MMLU-Pro leaderboard (TIGER-Lab), and Hugging Face Open LLM Leaderboard v2 (archived). Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.