Gemma 2 27B IT — Benchmarks

Benchmark scores for Gemma 2 27B IT aggregated from public leaderboards (6 boards), with how it ranks among open models and how it compares with its nearest neighbours. See hardware requirements for what you need to run it.

llmrun Composite Score →

72.7

90% range 66.2–79.2 · from 3 boards

CodingNot enough data yet
AgentsNot enough data yet
MathNot enough data yet
ScienceNot enough data yet
ReasoningNot enough data yet

Gemma 2 27B IT reasoning benchmarks

Multi-step logic and problem-solving — can the model think through a hard problem rather than recall an answer.

GPQA Diamond
GPQA Diamond is a set of extremely hard, graduate-level science questions (physics, chemistry, biology) written by domain experts and filtered so that skilled non-experts with web access still fail. It measures genuine reasoning rather than memorization.
36.5%
36.5%
median 54.0%open leader 93.1%
Overall best: Gemini 4 Argon (high) 96.4%
25th percentile · #74 of 98 open · #263 of 300 all
#74 of 98
#263 of 300
Independent
Epoch AI · Jun 2024

Gemma 2 27B IT math benchmarks

Symbolic and quantitative problem-solving, from competition math to multi-step calculation.

AIME 2024/2025
AIME (American Invitational Mathematics Examination) is a prestigious high-school competition of hard, integer-answer problems. It's a widely-cited yardstick for multi-step mathematical reasoning, here on the 2024–2025 papers.
1.4%
1.4%
median 52.8%open leader 98.6%
Overall best: Claude Fable 5 (high) 100.0%
9th percentile · #78 of 86 open · #272 of 283 all
#78 of 86
#272 of 283
Independent
Epoch AI · Jun 2024
MATH Level 5
MATH Level 5 covers the hardest tier of competition-mathematics problems, testing multi-step symbolic and quantitative reasoning.
27.9%
27.9%
median 41.6%open leader 96.6%
Overall best: GPT 5 (Aug 07, 2025, high) 98.1%
39th percentile · #24 of 39 open · #79 of 107 all
#24 of 39
#79 of 107
Independent
Epoch AI · Jun 2024

Gemma 2 27B IT knowledge benchmarks

Breadth and depth of factual knowledge across many subjects, paired with the reasoning to apply it.

MMLU-Pro
MMLU-Pro is a harder, cleaned-up successor to MMLU with ten answer choices and more reasoning-heavy questions across 14 subjects, measuring broad knowledge and reasoning together.
56.5%
56.5%
median 52.1%open leader 88.0%
Overall best: Gemini-3.1-Pro 91.2%
58th percentile · #62 of 145 open · #149 of 257 all
#62 of 145
#149 of 257
Independent
TIGER-Lab

Other results (1)· leaderboard-only, not part of the llmrun Score

MMLU
MMLU (Massive Multitask Language Understanding) spans 57 subjects from history to law to medicine as multiple-choice questions. It is the long-standing default for broad knowledge and remains the most widely-reported general benchmark.
75.7%
75.7%
median 62.5%open leader 87.2%
Overall best: GPT 4o (Nov 20, 2024) 88.1%
78th percentile · #19 of 85 open · #42 of 136 all
#19 of 85
#42 of 136
Lab-reported
Epoch AI · Jun 2024

Gemma 2 27B IT instruction following benchmarks

How precisely a model obeys explicit, verifiable instructions — format, length, and content constraints — and, more broadly, how much people prefer its responses head-to-head in blind human voting.

LMArena Text
Arena (arena.ai, formerly LMArena / Chatbot Arena) ranks models by millions of blind, head-to-head human votes on real prompts. The Arena Score shown here is a Bradley-Terry rating computed with Arena's default style control, so higher means people preferred that model's answers more often — it's a relative rating, not a percentage, so compare it only to other models on this board.
1289.5
1289.5
median 1321.5open leader 1487.1
Overall best: Gemini 4 Argon (High) 1523.9
40th percentile · #112 of 187 open · #283 of 405 all
#112 of 187
#283 of 405
Arena
Arena · Oct 2026

Score vs VRAM in the 24 GB tier

Gemma 2 27B IT
Other models
Best Score for its VRAM
x: VRAM (GB) · y: llmrun Score

#28 of 35 that fit 24 GB

Each dot is a model at the best quant that fits a 24 GB GPU. The line joins the models with the best Score for their VRAM; a dot below it is beaten by a model that needs no more memory.

Full 24 GB ranking →

Gemma 2 27B IT vs its neighbours

Same boards for the nearest open models per llmrun Score and the best model that fits 24 GB. The highest score in each row is marked.

Board	Gemma 2 27B ITthis model	Qwen3.8 27Bbest in classQ4_K_M · 18 GB	Gemma 3 27B ITnearest by ScoreQ4_K_M · 22 GB	Mistral Small 3.1 24B Instruct 2503nearest by ScoreQ4_K_M · 18 GB	Gemma 3 12B ITnearest by ScoreQ4_K_M · 11 GB
llmrun Score	72.7	122.1	79.4	79.0	74.2
GPQA Diamond	36.5%	—	47.7%	47.5%	39.5%
AIME 2024/2025	1.4%	—	22.5%	5.8%	16.7%
MATH Level 5	27.9%	—	74.0%	46.8%	—
MMLU-Pro	56.5%	—	67.5%	—	60.6%
LMArena Text	1289.5	1437.8	1365.5	1302.7	1341.8
Compare head to head		Compare	Compare	Compare	Compare

Scores aggregated from public benchmark sources (Epoch AI: 4 boards, Arena: 1 board, TIGER-Lab: 1 board) — each linked from the benchmark pages. llmrun does not run these benchmarks. Percentiles and ranks are among open models in the llmrun catalogue; the median and leader marks refer to the same field.