Noometry

Model comparison

Claude 3 Opus vs gpt-oss-20b

gpt-oss-20b is the stronger model overall, scoring 32.1 to 29.3 on the Noometry Index.

Last verified . 23 shared benchmarks.

Claude 3 Opus Anthropic

29.3

Rank #300 Confirmed

gpt-oss-20b OpenAI

32.1

Rank #249 Confirmed

Summary

  • They share 23 benchmarks with published results for both. Claude 3 Opus scores higher in 4 categories and gpt-oss-20b in 5 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where gpt-oss-20b leads 37.6 to 13.9.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 65.3% for gpt-oss-20b.
  • gpt-oss-20b has downloadable open weights; the other is API-only.

Side by side

Claude 3 Opusgpt-oss-20b
ProviderAnthropicOpenAI
Noometry Index29.332.1
Released2024-02-292025-08-05
WeightsProprietaryOpen
Context window—131K
Max output—16K
Input $ / M tokens—$0.018
Output $ / M tokens—$0.09
Results tracked4634

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

Category by category

Coding gpt-oss-20b leads

Claude 3 Opus: 32.4 (#261), gpt-oss-20b: 37.1 (#185)

BenchmarkClaude 3 Opusgpt-oss-20b
WeirdML19.2%40.9%
LMArena Coding12641307
SciCode—34.4%
BigCodeBench Instruct45.5%—
LiveBench Coding38.6%—
BigCodeBench Complete57.4%—
ALE-Bench—566.05
HumanEval+77.4%—
MBPP+73.3%—

Agentic & Tool Use Claude 3 Opus leads

Claude 3 Opus: 24.3 (#110), gpt-oss-20b: 9.2 (#146)

BenchmarkClaude 3 Opusgpt-oss-20b
Terminal-Bench—3.4%
Cybench10%—
METR Time Horizons29.5%—

Reasoning gpt-oss-20b leads

Claude 3 Opus: 14.2 (#318), gpt-oss-20b: 18.8 (#251)

BenchmarkClaude 3 Opusgpt-oss-20b
Chess Puzzles5%4%
LMArena Hard Prompts12461274
DTBench61.6%68%
LMCA17%14.5%
Epoch Capabilities Index126.91137.81
SimpleBench23.5%—
Kagi LLM Benchmark—53.2%
CritPt—1.4%
EnigmaEval0.8%—
LiveBench Reasoning40.6%—
LiveBench Data Analysis57.9%—
ForecastBench58.4—
LiveBench49.2%—
WinoGrande88.5%—

Math gpt-oss-20b leads

Claude 3 Opus: 13.9 (#293), gpt-oss-20b: 37.6 (#93)

BenchmarkClaude 3 Opusgpt-oss-20b
OTIS Mock AIME 2024-20254.7%65.3%
LMArena Math12731317
Omni-MATH—56.5%
LiveBench Math43.6%—
MATH Level 537.5%—

Knowledge gpt-oss-20b leads

Claude 3 Opus: 24.5 (#262), gpt-oss-20b: 34.6 (#186)

BenchmarkClaude 3 Opusgpt-oss-20b
GPQA Diamond47.2%60.8%
LMArena Expert12231258
SimpleQA Verified12.6%—
MMLU-Pro—74%
Confabulations22.7%—
GPQA (HELM)—59.4%
MMLU84.6%—

Multimodal Not comparable

Claude 3 Opus: 26.6 (#112), gpt-oss-20b: —

BenchmarkClaude 3 Opusgpt-oss-20b
LMArena Vision1023—

Multilingual Too close to call

Claude 3 Opus: 41.6 (#202), gpt-oss-20b: 42.4 (#192)

BenchmarkClaude 3 Opusgpt-oss-20b
LMArena Non-English12581269
LMArena Chinese12481311
LMArena German12581249
LMArena Japanese12041245
LMArena Korean11871239
LMArena Russian12801279
LMArena Spanish12461267
LMArena French1274—

Instruction Following Claude 3 Opus leads

Claude 3 Opus: 64.3 (#223), gpt-oss-20b: 62.0 (#235)

BenchmarkClaude 3 Opusgpt-oss-20b
LMArena Instruction Following12481236
LiveBench Instruction Following63.9%—
IFEval—73.2%

Long Context Too close to call

Claude 3 Opus: 38.2 (#197), gpt-oss-20b: 37.9 (#203)

BenchmarkClaude 3 Opusgpt-oss-20b
LMArena Longer Query12591251

Writing & Preference Claude 3 Opus leads

Claude 3 Opus: 47.4 (#208), gpt-oss-20b: 35.8 (#260)

BenchmarkClaude 3 Opusgpt-oss-20b
LMArena Text12621288
LMArena Creative Writing12351201
LMArena Multi-Turn12751268
EQ-Bench Creative Writing—666
WildBench—73.7%
LiveBench Language50.4%—

Frequently asked questions

Is Claude 3 Opus better than gpt-oss-20b?

gpt-oss-20b is the stronger model overall, scoring 32.1 to 29.3 on the Noometry Index.

Is Claude 3 Opus or gpt-oss-20b better for coding?

gpt-oss-20b scores higher on coding benchmarks: 37.1 versus 32.4 in the Noometry coding category.

How many benchmarks do Claude 3 Opus and gpt-oss-20b share?

23 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and gpt-oss-20b has 34.

Related comparisons

Go deeper