Noometry

Google, open weights

Gemma 2 27B

Gemma 2 27B by Google ranks 312th of 354 ranked models on the Noometry Index as of October 2026, with a score of 29.4. Its strongest category is long context, where it ranks 218th. API pricing starts at $0.65 per million input tokens and $0.65 per million output tokens, with a 8K-token context window.

Last verified

Noometry rank
#312 of 354
Index score
29.4
Evidence
Confirmed 34 results
Provider
Google
Released
June 24, 2024
Weights
Open weights
Reasoning
Unknown
Context window
8K
Max output
2K
Input price
$0.65 / M
Output price
$0.65 / M
Blended price
$0.65 / M
Output speed
Not measured
Value
#105 of 219
Knowledge cutoff
Unknown
Input
text
Hugging Face
google/gemma-2-27b-it

Category scores

Each category score combines every public result we have in that category.

  1. Coding 34.1
  2. Reasoning 15.3
  3. Math 10.7
  4. Knowledge 19.0
  5. Multilingual 38.6
  6. Instruction Following 60.5
  7. Long Context 37.3
  8. Writing & Preference 44.2
CategoryScoreRankResults
Coding34.1#2464
Reasoning15.3#3155
Math10.7#3114
Knowledge19.0#2803
Multilingual38.6#2261
Instruction Following60.5#2492
Long Context37.3#2181
Writing & Preference44.2#2254

Strengths and weaknesses

Categories where Gemma 2 27B places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Writing & Preference44.2−9.6#225 of 312, top 73%
Coding34.1−4.7#246 of 340, top 73%
Long Context37.3−3.6#218 of 296, top 74%

Weakest categories

CategoryScorevs medianRank
Math10.7−25.9#311 of 327, top 96%
Reasoning15.3−8.3#315 of 350, top 90%
Knowledge19.0−18.3#280 of 314, top 90%

Closest competitors

The models ranked just above and below Gemma 2 27B. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Llama 3.1-70B#30829.6$0.40—Compare
Llama 2-13B#30929.6——Compare
Claude 3 Opus#31029.5——Compare
DBRX#31129.4——Compare
Gemma 1.1 2b IT#31329.3——Compare
Phi 3 Small 8k Instruct#31429.3——Compare
Claude 3.5 Haiku#31529.2——Compare
GPT-4#31629.1$37.50—Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
BigCodeBench Instruct42.8%#27 of 64, top 43%BigCodeBench2024-06-19
LiveBench Coding36%#28 of 39, top 72%Epoch AI
LMArena Coding1211#236 of 294, top 81%LMArena2026-10-08
BigCodeBench Complete52.5%#27 of 66, top 41%BigCodeBench2024-06-19

Reasoning

BenchmarkScorePositionSettingSourceDate
LiveBench Reasoning28.1%#32 of 39, top 83%Epoch AI
LMArena Hard Prompts1198#237 of 297, top 80%LMArena2026-10-08
DTBench48%#142 of 151, top 95%Epoch AI
LiveBench Data Analysis47.9%#28 of 39, top 72%Epoch AI
LMCA7.1%#118 of 125, top 95%Epoch AI
Epoch Capabilities Index122.08#162 of 213, top 77%Epoch AI2024-06-24
LiveBench38.2%#31 of 39, top 80%Epoch AI

Math

BenchmarkScorePositionSettingSourceDate
OTIS Mock AIME 2024-20251.4%#165 of 173, top 96%Epoch AI2025-03-07
LiveBench Math26.5%#33 of 39, top 85%Epoch AI
LMArena Math1212#228 of 285, top 80%LMArena2026-10-08
MATH Level 527.9%#56 of 79, top 71%Epoch AI2025-01-27

Knowledge

BenchmarkScorePositionSettingSourceDate
GPQA Diamond36.5%#155 of 186, top 84%Epoch AI2025-01-27
Confabulations (lower is better)27.1%#44 of 51, top 87%Lech Mazur benchmarks
LMArena Expert1172#229 of 273, top 84%LMArena2026-10-08
MMLU75.7%#32 of 81, top 40%Epoch AI

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1217#226 of 297, top 77%LMArena2026-10-08
LMArena Chinese1221#223 of 285, top 79%LMArena2026-10-08
LMArena French1247#184 of 223, top 83%LMArena2026-10-08
LMArena German1209#189 of 231, top 82%LMArena2026-10-08
LMArena Japanese1175#166 of 211, top 79%LMArena2026-10-08
LMArena Korean1174#172 of 213, top 81%LMArena2026-10-08
LMArena Russian1234#220 of 283, top 78%LMArena2026-10-08
LMArena Spanish1228#189 of 226, top 84%LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LiveBench Instruction Following58.1%#30 of 39, top 77%Epoch AI
LMArena Instruction Following1206#231 of 298, top 78%LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1231#229 of 291, top 79%LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1231#229 of 297, top 78%LMArena2026-10-08
LMArena Creative Writing1241#215 of 295, top 73%LMArena2026-10-08
LMArena Multi-Turn1224#229 of 295, top 78%LMArena2026-10-08
LiveBench Language32.6%#25 of 39, top 65%Epoch AI

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
openrouter$0.65$0.65—2026-10-10

All Google API prices → Estimate your cost →

Compare Gemma 2 27B

Other Google models

Frequently asked questions

How good is Gemma 2 27B?

Gemma 2 27B by Google ranks 312th of 354 ranked models on the Noometry Index as of October 2026, with a score of 29.4. Its strongest category is long context, where it ranks 218th. API pricing starts at $0.65 per million input tokens and $0.65 per million output tokens, with a 8K-token context window.

How much does Gemma 2 27B cost?

Gemma 2 27B costs $0.65 per million input tokens and $0.65 per million output tokens on openrouter.

What is Gemma 2 27B's context window?

Gemma 2 27B accepts up to 8K tokens of input and can write up to 2K tokens in one response.

Is Gemma 2 27B open source?

Yes. Gemma 2 27B's weights are downloadable from Hugging Face (google/gemma-2-27b-it); check the license for commercial terms.

What are Gemma 2 27B's strengths and weaknesses?

Relative to other ranked models, Gemma 2 27B places best in writing & preference, coding, long context and lowest in math, reasoning, knowledge.

What is Gemma 2 27B best at?

Its best category is long context, where it ranks 218th on Noometry.