Noometry

Model comparison

Claude Opus 4.5 vs gpt-oss-120b

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 36.3 on the Noometry Index. gpt-oss-120b costs 142× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

Last verified . 35 shared benchmarks.

Claude Opus 4.5 Anthropic

50.5

Rank #47 Confirmed

gpt-oss-120b OpenAI

36.3

Rank #217 Confirmed

Summary

  • They share 35 benchmarks with published results for both. Claude Opus 4.5 scores higher in 8 categories and gpt-oss-120b in 1 category; 9 gaps are clear of the uncertainty.
  • The widest gap is in agentic & tool use, where Claude Opus 4.5 leads 47.3 to 12.2.
  • The biggest single-benchmark swing is SWE-bench Verified (bash only): 76.8% for Claude Opus 4.5 and 26% for gpt-oss-120b.
  • gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
  • Claude Opus 4.5 accepts more context: 200K tokens versus 131K.
  • gpt-oss-120b has downloadable open weights; the other is API-only.

Side by side

Claude Opus 4.5gpt-oss-120b
ProviderAnthropicOpenAI
Noometry Index50.536.3
Released2025-11-012025-08-05
WeightsProprietaryOpen
Context window200K131K
Max output64K41K
Input $ / M tokens$5$0.037
Output $ / M tokens$25$0.17
Results tracked6948

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

Category by category

Coding Claude Opus 4.5 leads

Claude Opus 4.5: 54.8 (#27), gpt-oss-120b: 33.5 (#256)

BenchmarkClaude Opus 4.5gpt-oss-120b
SWE-bench Verified (bash only)76.8%26%
WeirdML63.7%48.2%
LMArena Coding15041380
ALE-Bench1,025575.62
AlgoTune1.771.41
SWE-bench Verified76.7%—
Aider Polyglot—41.8%
LMArena WebDev1494—
SWE-bench Multilingual70.7%—
SciCode—36%
GSO26.5%—

Agentic & Tool Use Claude Opus 4.5 leads

Claude Opus 4.5: 47.3 (#12), gpt-oss-120b: 12.2 (#153)

BenchmarkClaude Opus 4.5gpt-oss-120b
Terminal-Bench63.1%18.7%
METR Time Horizons75%56.6%
Vending-Bench 24,967-21.53
APEX-Agents—4.4%
Berkeley Function Calling Leaderboard77.5%—
GDPval45.5%—
Remote Labor Index3.8%—
τ²-bench Airline84%—
τ²-bench Banking24.7%—
τ²-bench Retail79.6%—
τ²-bench Telecom92.3%—
Cybench82%—
DeepResearch Bench54.8%—
OSWorld66.3%—
BALROG43.5%—
LMArena Search1180—

Reasoning Claude Opus 4.5 leads

Claude Opus 4.5: 42.6 (#51), gpt-oss-120b: 20.0 (#245)

BenchmarkClaude Opus 4.5gpt-oss-120b
SimpleBench62%22.1%
Kagi LLM Benchmark80.2%58.6%
Chess Puzzles12%20%
LMArena Hard Prompts14761364
Mystery Game Puzzles22%2%
DTBench89.9%76.3%
LMCA44.5%22.1%
Epoch Capabilities Index150.09139.93
ARC-AGI-237.6%—
NYT Connections (extended)52.5%—
ARC-AGI-180%—
CritPt—1.1%
EnigmaEval11.9%—
EBR-Bench14.3%—
Surface Evolver Bench—25%
ForecastBench60.7—

Math gpt-oss-120b leads

Claude Opus 4.5: 38.6 (#132), gpt-oss-120b: 52.5 (#50)

BenchmarkClaude Opus 4.5gpt-oss-120b
OTIS Mock AIME 2024-202586.1%88.9%
LMArena Math14631389
FrontierMath (Tiers 1-3)34.4%—
FrontierMath Tier 44.9%—
ProofBench36%—
Omni-MATH—68.8%
FrontierMath (Feb 2025 set)20.7%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge Claude Opus 4.5 leads

Claude Opus 4.5: 56.5 (#44), gpt-oss-120b: 42.4 (#96)

BenchmarkClaude Opus 4.5gpt-oss-120b
GPQA Diamond86%75.8%
Vectara Hallucination Rate10.9%14.2%
LMArena Expert14871356
Humanity's Last Exam25.2%—
SimpleQA Verified45.7%—
MMLU-Pro—79.5%
Confabulations—15.7%
GPQA (HELM)—68.4%

Multimodal Not comparable

Claude Opus 4.5: 31.4 (#107), gpt-oss-120b: —

BenchmarkClaude Opus 4.5gpt-oss-120b
GeoBench75%—
VPCT40%—
Furniture Assembly28.3%—
LMArena Document1462—

Multilingual Claude Opus 4.5 leads

Claude Opus 4.5: 54.3 (#47), gpt-oss-120b: 48.0 (#147)

BenchmarkClaude Opus 4.5gpt-oss-120b
LMArena Non-English14381351
LMArena Chinese14701385
LMArena French14711369
LMArena German14491353
LMArena Japanese14161331
LMArena Korean14241282
LMArena Russian14471343
LMArena Spanish14581389

Instruction Following Claude Opus 4.5 leads

Claude Opus 4.5: 77.5 (#19), gpt-oss-120b: 69.3 (#173)

BenchmarkClaude Opus 4.5gpt-oss-120b
LMArena Instruction Following14781318
IFEval—83.6%

Long Context Claude Opus 4.5 leads

Claude Opus 4.5: 46.5 (#22), gpt-oss-120b: 31.4 (#278)

BenchmarkClaude Opus 4.5gpt-oss-120b
LMArena Longer Query14801319
Fiction.LiveBench—44.4%
CL-bench21.1%—

Writing & Preference Claude Opus 4.5 leads

Claude Opus 4.5: 68.1 (#28), gpt-oss-120b: 46.5 (#217)

BenchmarkClaude Opus 4.5gpt-oss-120b
LMArena Text14511365
LMArena Creative Writing14451275
EQ-Bench Creative Writing1687961
LMArena Multi-Turn14661340
Short-Story Creative Writing—77.1%
WildBench—84.5%

Frequently asked questions

Is Claude Opus 4.5 better than gpt-oss-120b?

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 36.3 on the Noometry Index. gpt-oss-120b costs 142× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 4.5 or gpt-oss-120b?

gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Claude Opus 4.5 lists at $5 and $25.

Is Claude Opus 4.5 or gpt-oss-120b better for coding?

Claude Opus 4.5 scores higher on coding benchmarks: 54.8 versus 33.5 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 4.5 does, with 200K tokens against 131K.

How many benchmarks do Claude Opus 4.5 and gpt-oss-120b share?

35 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and gpt-oss-120b has 48.

Related comparisons

Go deeper