Model comparison

Gemma 2 27B vs phi-3-medium 14B

Gemma 2 27B and phi-3-medium 14B score almost the same on the Noometry Index (29.4 vs 29.7), so choose on price, context window or the category you care about most.

Last verified . 6 shared benchmarks.

Gemma 2 27B Google

29.4

Rank #312 Confirmed

phi-3-medium 14B Microsoft

29.7

Rank #306 Reported

Summary

  • They share 6 benchmarks with published results for both. Gemma 2 27B scores higher in 1 category and phi-3-medium 14B in 2 categories; 3 gaps are clear of the uncertainty.
  • The widest gap is in math, where phi-3-medium 14B leads 27.3 to 10.7.
  • The biggest single-benchmark swing is MATH Level 5: 27.9% for Gemma 2 27B and 17.6% for phi-3-medium 14B.

Side by side

Gemma 2 27B and phi-3-medium 14B specifications
Gemma 2 27Bphi-3-medium 14B
ProviderGoogleMicrosoft
Noometry Index29.429.7
Released2024-06-242024-04-23
WeightsOpenOpen
Context window8K—
Max output2K—
Input $ / M tokens$0.65—
Output $ / M tokens$0.65—
Results tracked3413

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

Category by category

Coding phi-3-medium 14B leads

Gemma 2 27B: 34.1 (#246), phi-3-medium 14B: 36.8 (#201)

Coding benchmarks
BenchmarkGemma 2 27Bphi-3-medium 14B
BigCodeBench Instruct42.8%37.6%
BigCodeBench Complete52.5%48.7%
LiveBench Coding36%—
LMArena Coding1211—

Reasoning Not comparable

Gemma 2 27B: 15.3 (#315), phi-3-medium 14B: —

Reasoning benchmarks
BenchmarkGemma 2 27Bphi-3-medium 14B
Epoch Capabilities Index122.08121.23
LiveBench Reasoning28.1%—
LMArena Hard Prompts1198—
DTBench48%—
LiveBench Data Analysis47.9%—
LMCA7.1%—
Adversarial NLI—55.8%
BIG-Bench Hard—81.4%
HellaSwag—82.4%
LiveBench38.2%—
WinoGrande—81.5%

Math phi-3-medium 14B leads

Gemma 2 27B: 10.7 (#311), phi-3-medium 14B: 27.3 (#250)

Math benchmarks
BenchmarkGemma 2 27Bphi-3-medium 14B
MATH Level 527.9%17.6%
OTIS Mock AIME 2024-20251.4%—
LiveBench Math26.5%—
LMArena Math1212—

Knowledge Gemma 2 27B leads

Gemma 2 27B: 19.0 (#280), phi-3-medium 14B: 9.1 (#306)

Knowledge benchmarks
BenchmarkGemma 2 27Bphi-3-medium 14B
GPQA Diamond36.5%27.6%
MMLU75.7%78%
Confabulations27.1%—
LMArena Expert1172—
ARC (AI2) Challenge—91.6%
OpenBookQA—87.4%
TriviaQA—73.9%

Multilingual Not comparable

Gemma 2 27B: 38.6 (#226), phi-3-medium 14B: —

Multilingual benchmarks
BenchmarkGemma 2 27Bphi-3-medium 14B
LMArena Non-English1217—
LMArena Chinese1221—
LMArena French1247—
LMArena German1209—
LMArena Japanese1175—
LMArena Korean1174—
LMArena Russian1234—
LMArena Spanish1228—

Instruction Following Not comparable

Gemma 2 27B: 60.5 (#249), phi-3-medium 14B: —

Instruction Following benchmarks
BenchmarkGemma 2 27Bphi-3-medium 14B
LiveBench Instruction Following58.1%—
LMArena Instruction Following1206—

Long Context Not comparable

Gemma 2 27B: 37.3 (#218), phi-3-medium 14B: —

Long Context benchmarks
BenchmarkGemma 2 27Bphi-3-medium 14B
LMArena Longer Query1231—

Writing & Preference Not comparable

Gemma 2 27B: 44.2 (#225), phi-3-medium 14B: —

Writing & Preference benchmarks
BenchmarkGemma 2 27Bphi-3-medium 14B
LMArena Text1231—
LMArena Creative Writing1241—
LMArena Multi-Turn1224—
LiveBench Language32.6%—

Frequently asked questions

Is Gemma 2 27B better than phi-3-medium 14B?

Gemma 2 27B and phi-3-medium 14B score almost the same on the Noometry Index (29.4 vs 29.7), so choose on price, context window or the category you care about most.

Is Gemma 2 27B or phi-3-medium 14B better for coding?

phi-3-medium 14B scores higher on coding benchmarks: 36.8 versus 34.1 in the Noometry coding category.

How many benchmarks do Gemma 2 27B and phi-3-medium 14B share?

6 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and phi-3-medium 14B has 13.

Related comparisons

Go deeper