Noometry

Model comparison

Claude Opus 5 vs Gemini 2.5 Pro

Claude Opus 5 is the stronger model overall, scoring 67.8 to 45.0 on the Noometry Index. Gemini 2.5 Pro costs 2.9× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.

Last verified . 39 shared benchmarks.

Claude Opus 5 Anthropic

67.8

Rank #4 Confirmed

Gemini 2.5 Pro Google

45.0

Rank #75 Confirmed

Summary

  • They share 39 benchmarks with published results for both. Claude Opus 5 scores higher in 9 categories and Gemini 2.5 Pro in 1 category; 10 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Opus 5 leads 86.2 to 32.5.
  • The biggest single-benchmark swing is ARC-AGI-2: 90.4% for Claude Opus 5 and 4.9% for Gemini 2.5 Pro.
  • Gemini 2.5 Pro is cheaper at $1.25 / $10 per million input/output tokens, against $5 / $25 for Claude Opus 5.
  • Gemini 2.5 Pro accepts more context: 1.05M tokens versus 1M.

Side by side

Claude Opus 5Gemini 2.5 Pro
ProviderAnthropicGoogle
Noometry Index67.845.0
Released2026-07-242025-03-25
WeightsProprietaryProprietary
Context window1M1.05M
Max output128K66K
Input $ / M tokens$5$1.25
Output $ / M tokens$25$10
Results tracked5778

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

Category by category

Coding Claude Opus 5 leads

Claude Opus 5: 67.5 (#5), Gemini 2.5 Pro: 42.4 (#101)

BenchmarkClaude Opus 5Gemini 2.5 Pro
LMArena WebDev16911227
SciCode56.4%42.8%
WeirdML91.8%54%
LMArena Coding15341452
ALE-Bench2,165785.52
SWE-bench Verified—57.6%
DeepSWE73.6%—
FrontierCode53.4%—
SWE-bench Verified (bash only)—53.6%
Aider Polyglot—83.1%
CursorBench46.6%—
FrontierSWE52%—
GSO—3.9%
LiveBench Coding—85.9%
CadEval—64%
AlgoTune—1.51

Agentic & Tool Use Claude Opus 5 leads

Claude Opus 5: 55.6 (#1), Gemini 2.5 Pro: 29.2 (#88)

BenchmarkClaude Opus 5Gemini 2.5 Pro
τ²-bench Banking48.7%13.7%
BALROG63.4%43.3%
Vending-Bench 211,182573.64
Terminal-Bench—32.6%
APEX-Agents65.8%—
OSWorld 2.031.4%—
GDPval—23.3%
Remote Labor Index—0.8%
TheAgentCompany—30.3%
DeepResearch Bench—42.8%
PostTrainBench35%—
GBAEval79.6%—
GDP.pdf24%—
LMArena Search—1142
METR Time Horizons—55.4%

Reasoning Claude Opus 5 leads

Claude Opus 5: 77.2 (#4), Gemini 2.5 Pro: 28.8 (#99)

BenchmarkClaude Opus 5Gemini 2.5 Pro
ARC-AGI-290.4%4.9%
SimpleBench80.6%62.4%
ARC-AGI-197.5%41%
CritPt29.1%2%
Chess Puzzles42%20%
LMArena Hard Prompts15261455
DTBench97.9%82.4%
LMCA64.5%34.8%
Epoch Capabilities Index162.78145.32
Kagi LLM Benchmark—70.3%
NYT Connections (extended)94.3%—
EnigmaEval—5.6%
EBR-Bench45.7%—
LiveBench Reasoning—89.8%
Mystery Game Puzzles59%—
LiveBench Data Analysis—79.9%
Bench to the Future 30.12—
ForecastBench—61.3
LiveBench—82.3%

Math Claude Opus 5 leads

Claude Opus 5: 86.2 (#8), Gemini 2.5 Pro: 32.5 (#213)

BenchmarkClaude Opus 5Gemini 2.5 Pro
FrontierMath (Tiers 1-3)85.6%24.6%
FrontierMath Tier 473.2%0%
OTIS Mock AIME 2024-202598.9%84.7%
LMArena Math15311450
ProofBench99%—
Omni-MATH—41.6%
LiveBench Math—90.2%
MATH Level 5—95.9%
FrontierMath (Feb 2025 set)—14.1%
FrontierMath Tier 4 (v1)—4.2%

Knowledge Claude Opus 5 leads

Claude Opus 5: 66.8 (#9), Gemini 2.5 Pro: 56.0 (#46)

BenchmarkClaude Opus 5Gemini 2.5 Pro
GPQA Diamond93.9%85.3%
LMArena Expert15571452
Humanity's Last Exam—21.6%
SimpleQA Verified59.9%—
MMLU-Pro—86.3%
Confabulations—10.6%
Vectara Hallucination Rate—7%
GPQA (HELM)—74.9%

Multimodal Claude Opus 5 leads

Claude Opus 5: 50.8 (#8), Gemini 2.5 Pro: 45.2 (#18)

BenchmarkClaude Opus 5Gemini 2.5 Pro
LMArena Vision13191263
LMArena Document15161421
GeoBench—86%
VPCT—48%
Blueprint-Bench 230.4%—
Furniture Assembly60.8%—
SpatialViz-Bench—44.7%

Multilingual Claude Opus 5 leads

Claude Opus 5: 58.8 (#4), Gemini 2.5 Pro: 55.3 (#31)

BenchmarkClaude Opus 5Gemini 2.5 Pro
LMArena Non-English15011451
LMArena Chinese15741507
LMArena French15191472
LMArena German15241487
LMArena Japanese15161461
LMArena Korean15211434
LMArena Russian15071461
LMArena Spanish15191473

Instruction Following Claude Opus 5 leads

Claude Opus 5: 79.2 (#7), Gemini 2.5 Pro: 75.0 (#75)

BenchmarkClaude Opus 5Gemini 2.5 Pro
LMArena Instruction Following15171437
LiveBench Instruction Following—80.6%
IFEval—84%

Long Context Gemini 2.5 Pro leads

Claude Opus 5: 46.5 (#21), Gemini 2.5 Pro: 59.8 (#5)

BenchmarkClaude Opus 5Gemini 2.5 Pro
LMArena Longer Query15151449
Fiction.LiveBench—91.7%

Writing & Preference Claude Opus 5 leads

Claude Opus 5: 79.2 (#1), Gemini 2.5 Pro: 63.7 (#62)

BenchmarkClaude Opus 5Gemini 2.5 Pro
LMArena Text15071458
LMArena Creative Writing14911454
EQ-Bench Creative Writing21331421
LMArena Multi-Turn14991453
Short-Story Creative Writing—83.8%
WildBench—85.7%
EQ-Bench 41385—
LiveBench Language—67.8%

Frequently asked questions

Is Claude Opus 5 better than Gemini 2.5 Pro?

Claude Opus 5 is the stronger model overall, scoring 67.8 to 45.0 on the Noometry Index. Gemini 2.5 Pro costs 2.9× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 5 or Gemini 2.5 Pro?

Gemini 2.5 Pro is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Claude Opus 5 lists at $5 and $25.

Is Claude Opus 5 or Gemini 2.5 Pro better for coding?

Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 42.4 in the Noometry coding category.

Which has the bigger context window?

Gemini 2.5 Pro does, with 1.05M tokens against 1M.

How many benchmarks do Claude Opus 5 and Gemini 2.5 Pro share?

39 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Gemini 2.5 Pro has 78.

Related comparisons

Go deeper