Noometry

Model comparison

GPT-4 vs phi-3-medium 14B

GPT-4 and phi-3-medium 14B score almost the same on the Noometry Index (29.1 vs 29.7), so choose on price, context window or the category you care about most.

Last verified . 10 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

phi-3-medium 14B Microsoft

29.7

Rank #306 Reported

Summary

  • They share 10 benchmarks with published results for both. GPT-4 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 10.8.
  • The biggest single-benchmark swing is BigCodeBench Complete: 57.2% for GPT-4 and 48.7% for phi-3-medium 14B.
  • phi-3-medium 14B has downloadable open weights; the other is API-only.

Side by side

GPT-4phi-3-medium 14B
ProviderOpenAIMicrosoft
Noometry Index29.129.7
Released2023-03-142024-04-23
WeightsProprietaryOpen
Context window8K—
Max output8K—
Input $ / M tokens$30—
Output $ / M tokens$60—
Results tracked3813

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

Category by category

Coding phi-3-medium 14B leads

GPT-4: 31.6 (#283), phi-3-medium 14B: 36.8 (#201)

BenchmarkGPT-4phi-3-medium 14B
BigCodeBench Instruct46%37.6%
BigCodeBench Complete57.2%48.7%
WeirdML12.4%—
LMArena Coding1254—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, phi-3-medium 14B: —

BenchmarkGPT-4phi-3-medium 14B
METR Time Horizons36.1%—

Reasoning Not comparable

GPT-4: 17.8 (#289), phi-3-medium 14B: —

BenchmarkGPT-4phi-3-medium 14B
BIG-Bench Hard75.1%81.4%
Epoch Capabilities Index125.89121.23
HellaSwag95.3%82.4%
WinoGrande87.5%81.5%
Chess Puzzles4%—
LMArena Hard Prompts1241—
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
Adversarial NLI—55.8%
ForecastBench57.8—

Math phi-3-medium 14B leads

GPT-4: 10.8 (#309), phi-3-medium 14B: 27.3 (#250)

BenchmarkGPT-4phi-3-medium 14B
MATH Level 523%17.6%
OTIS Mock AIME 2024-20251.1%—
LMArena Math1269—
GSM8K92%—

Knowledge GPT-4 leads

GPT-4: 18.4 (#282), phi-3-medium 14B: 9.1 (#306)

BenchmarkGPT-4phi-3-medium 14B
GPQA Diamond35.7%27.6%
MMLU86.4%78%
TriviaQA84.8%73.9%
LMArena Expert1211—
ARC (AI2) Challenge—91.6%
OpenBookQA—87.4%

Multilingual Not comparable

GPT-4: 40.6 (#215), phi-3-medium 14B: —

BenchmarkGPT-4phi-3-medium 14B
LMArena Non-English1246—
LMArena Chinese1242—
LMArena French1283—
LMArena German1251—
LMArena Japanese1209—
LMArena Korean1184—
LMArena Russian1251—
LMArena Spanish1261—

Instruction Following Not comparable

GPT-4: 65.3 (#222), phi-3-medium 14B: —

BenchmarkGPT-4phi-3-medium 14B
LMArena Instruction Following1241—

Long Context Not comparable

GPT-4: 37.7 (#212), phi-3-medium 14B: —

BenchmarkGPT-4phi-3-medium 14B
LMArena Longer Query1244—

Writing & Preference Not comparable

GPT-4: 34.9 (#268), phi-3-medium 14B: —

BenchmarkGPT-4phi-3-medium 14B
LMArena Text1263—
LMArena Creative Writing1244—
EQ-Bench Creative Writing752—
LMArena Multi-Turn1257—

Frequently asked questions

Is GPT-4 better than phi-3-medium 14B?

GPT-4 and phi-3-medium 14B score almost the same on the Noometry Index (29.1 vs 29.7), so choose on price, context window or the category you care about most.

Is GPT-4 or phi-3-medium 14B better for coding?

phi-3-medium 14B scores higher on coding benchmarks: 36.8 versus 31.6 in the Noometry coding category.

How many benchmarks do GPT-4 and phi-3-medium 14B share?

10 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and phi-3-medium 14B has 13.

Related comparisons

Go deeper