Noometry

Model comparison

GPT-4.5 vs phi-3-medium 14B

GPT-4.5 is the stronger model overall, scoring 36.9 to 29.1 on the Noometry Index.

Last verified . 3 shared benchmarks.

GPT-4.5 OpenAI

36.9

Rank #200 Confirmed

phi-3-medium 14B Microsoft

29.1

Rank #306 Reported

Summary

  • They share 3 benchmarks with published results for both. GPT-4.5 scores higher in 3 categories and phi-3-medium 14B in 0 categories; 3 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where GPT-4.5 leads 32.4 to 9.1.
  • The biggest single-benchmark swing is MATH Level 5: 78.6% for GPT-4.5 and 17.6% for phi-3-medium 14B.
  • phi-3-medium 14B has downloadable open weights; the other is API-only.

Side by side

GPT-4.5phi-3-medium 14B
ProviderOpenAIMicrosoft
Noometry Index36.929.1
Released2025-02-272024-04-23
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked4213

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

Category by category

Coding GPT-4.5 leads

GPT-4.5: 41.8 (#102), phi-3-medium 14B: 36.3 (#198)

BenchmarkGPT-4.5phi-3-medium 14B
Aider Polyglot44.9%—
WeirdML39.4%—
BigCodeBench Instruct—37.6%
LiveBench Coding75.2%—
LMArena Coding1397—
BigCodeBench Complete—48.7%

Agentic & Tool Use Not comparable

GPT-4.5: 27.6 (#92), phi-3-medium 14B: —

BenchmarkGPT-4.5phi-3-medium 14B
Cybench17.5%—

Reasoning Not comparable

GPT-4.5: 14.1 (#320), phi-3-medium 14B: —

BenchmarkGPT-4.5phi-3-medium 14B
Epoch Capabilities Index136.74121.18
ARC-AGI-20.8%—
SimpleBench34.5%—
ARC-AGI-110.3%—
EnigmaEval3.2%—
LiveBench Reasoning71.1%—
LMArena Hard Prompts1404—
LiveBench Data Analysis64.3%—
Adversarial NLI—55.8%
BIG-Bench Hard—81.4%
ForecastBench61.7—
HellaSwag—82.4%
LiveBench69%—
WinoGrande—81.5%

Math GPT-4.5 leads

GPT-4.5: 31.0 (#200), phi-3-medium 14B: 25.2 (#249)

BenchmarkGPT-4.5phi-3-medium 14B
MATH Level 578.6%17.6%
OTIS Mock AIME 2024-202537.8%—
LiveBench Math69.3%—
LMArena Math1412—

Knowledge GPT-4.5 leads

GPT-4.5: 32.4 (#205), phi-3-medium 14B: 9.1 (#301)

BenchmarkGPT-4.5phi-3-medium 14B
GPQA Diamond68.7%27.6%
Humanity's Last Exam5.4%—
Confabulations13.6%—
LMArena Expert1394—
ARC (AI2) Challenge—91.6%
MMLU—78%
OpenBookQA—87.4%
TriviaQA—73.9%

Multimodal Not comparable

GPT-4.5: 37.0 (#68), phi-3-medium 14B: —

BenchmarkGPT-4.5phi-3-medium 14B
LMArena Vision1195—
VPCT45%—

Multilingual Not comparable

GPT-4.5: 52.7 (#78), phi-3-medium 14B: —

BenchmarkGPT-4.5phi-3-medium 14B
LMArena Non-English1413—
LMArena Chinese1421—
LMArena French1417—
LMArena German1457—
LMArena Japanese1416—
LMArena Korean1392—
LMArena Russian1419—

Instruction Following Not comparable

GPT-4.5: 72.8 (#129), phi-3-medium 14B: —

BenchmarkGPT-4.5phi-3-medium 14B
LiveBench Instruction Following72.3%—
LMArena Instruction Following1404—

Long Context Not comparable

GPT-4.5: 40.4 (#150), phi-3-medium 14B: —

BenchmarkGPT-4.5phi-3-medium 14B
Fiction.LiveBench63.9%—
LMArena Longer Query1406—

Writing & Preference Not comparable

GPT-4.5: 57.1 (#129), phi-3-medium 14B: —

BenchmarkGPT-4.5phi-3-medium 14B
LMArena Text1417—
LMArena Creative Writing1394—
Short-Story Creative Writing75.6%—
EQ-Bench Creative Writing1258—
LMArena Multi-Turn1444—
LiveBench Language61.5%—

Frequently asked questions

Is GPT-4.5 better than phi-3-medium 14B?

GPT-4.5 is the stronger model overall, scoring 36.9 to 29.1 on the Noometry Index.

Is GPT-4.5 or phi-3-medium 14B better for coding?

GPT-4.5 scores higher on coding benchmarks: 41.8 versus 36.3 in the Noometry coding category.

How many benchmarks do GPT-4.5 and phi-3-medium 14B share?

3 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and phi-3-medium 14B has 13.

Related comparisons

Go deeper