Noometry

Model comparison

GPT-4.5 vs gpt-oss-20b

GPT-4.5 is the stronger model overall, scoring 37.2 to 32.5 on the Noometry Index.

Last verified . 20 shared benchmarks.

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

gpt-oss-20b OpenAI

32.5

Rank #255 Confirmed

Summary

  • They share 20 benchmarks with published results for both. GPT-4.5 scores higher in 6 categories and gpt-oss-20b in 3 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where GPT-4.5 leads 56.9 to 35.5.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 65.3% for gpt-oss-20b.
  • gpt-oss-20b has downloadable open weights; the other is API-only.

Side by side

GPT-4.5gpt-oss-20b
ProviderOpenAIOpenAI
Noometry Index37.232.5
Released2025-02-272025-08-05
WeightsProprietaryOpen
Context window—131K
Max output—16K
Input $ / M tokens—$0.018
Output $ / M tokens—$0.09
Results tracked4234

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

Category by category

Coding GPT-4.5 leads

GPT-4.5: 42.2 (#109), gpt-oss-20b: 37.6 (#192)

BenchmarkGPT-4.5gpt-oss-20b
WeirdML39.4%40.9%
LMArena Coding13961306
Aider Polyglot44.9%—
SciCode—34.4%
LiveBench Coding75.2%—
ALE-Bench—566.05

Agentic & Tool Use GPT-4.5 leads

GPT-4.5: 27.9 (#97), gpt-oss-20b: 9.3 (#154)

BenchmarkGPT-4.5gpt-oss-20b
Terminal-Bench—3.4%
Cybench17.5%—

Reasoning gpt-oss-20b leads

GPT-4.5: 13.9 (#330), gpt-oss-20b: 19.3 (#261)

BenchmarkGPT-4.5gpt-oss-20b
LMArena Hard Prompts14031274
Epoch Capabilities Index136.74137.82
ARC-AGI-20.8%—
SimpleBench34.5%—
Kagi LLM Benchmark—53.2%
ARC-AGI-110.3%—
CritPt—1.4%
Chess Puzzles—4%
EnigmaEval3.2%—
LiveBench Reasoning71.1%—
DTBench—68%
LiveBench Data Analysis64.3%—
LMCA—14.5%
ForecastBench61.7—
LiveBench69%—

Math gpt-oss-20b leads

GPT-4.5: 32.6 (#211), gpt-oss-20b: 39.4 (#103)

BenchmarkGPT-4.5gpt-oss-20b
OTIS Mock AIME 2024-202537.8%65.3%
LMArena Math14121317
Omni-MATH—56.5%
LiveBench Math69.3%—
MATH Level 578.6%—

Knowledge gpt-oss-20b leads

GPT-4.5: 32.5 (#211), gpt-oss-20b: 34.6 (#195)

BenchmarkGPT-4.5gpt-oss-20b
GPQA Diamond68.7%60.8%
LMArena Expert13941258
Humanity's Last Exam5.4%—
MMLU-Pro—74%
Confabulations13.6%—
GPQA (HELM)—59.4%

Multimodal Not comparable

GPT-4.5: 37.6 (#71), gpt-oss-20b: —

BenchmarkGPT-4.5gpt-oss-20b
LMArena Vision1195—
VPCT45%—

Multilingual GPT-4.5 leads

GPT-4.5: 52.5 (#83), gpt-oss-20b: 42.2 (#197)

BenchmarkGPT-4.5gpt-oss-20b
LMArena Non-English14131268
LMArena Chinese14211314
LMArena German14571255
LMArena Japanese14161244
LMArena Korean13921236
LMArena Russian14191278
LMArena French1418—
LMArena Spanish—1267

Instruction Following GPT-4.5 leads

GPT-4.5: 72.6 (#134), gpt-oss-20b: 61.8 (#240)

BenchmarkGPT-4.5gpt-oss-20b
LMArena Instruction Following14041236
LiveBench Instruction Following72.3%—
IFEval—73.2%

Long Context GPT-4.5 leads

GPT-4.5: 40.4 (#155), gpt-oss-20b: 37.9 (#209)

BenchmarkGPT-4.5gpt-oss-20b
LMArena Longer Query14061250
Fiction.LiveBench63.9%—

Writing & Preference GPT-4.5 leads

GPT-4.5: 56.9 (#134), gpt-oss-20b: 35.5 (#265)

BenchmarkGPT-4.5gpt-oss-20b
LMArena Text14171287
LMArena Creative Writing13941201
EQ-Bench Creative Writing1258666
LMArena Multi-Turn14441268
Short-Story Creative Writing75.6%—
WildBench—73.7%
LiveBench Language61.5%—

Frequently asked questions

Is GPT-4.5 better than gpt-oss-20b?

GPT-4.5 is the stronger model overall, scoring 37.2 to 32.5 on the Noometry Index.

Is GPT-4.5 or gpt-oss-20b better for coding?

GPT-4.5 scores higher on coding benchmarks: 42.2 versus 37.6 in the Noometry coding category.

How many benchmarks do GPT-4.5 and gpt-oss-20b share?

20 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and gpt-oss-20b has 34.

Related comparisons

Go deeper