Noometry

Model comparison

Claude Opus 4.6 vs Gemma 3 27B

Claude Opus 4.6 is the stronger model overall, scoring 58.3 to 30.5 on the Noometry Index. Gemma 3 27B costs 100× less per token, which makes it the better buy when Claude Opus 4.6's lead doesn't matter for your workload.

Last verified . 27 shared benchmarks.

Claude Opus 4.6 Anthropic

58.3

Rank #18 Confirmed

Gemma 3 27B Google

30.5

Rank #277 Confirmed

Summary

  • They share 27 benchmarks with published results for both. Claude Opus 4.6 scores higher in 10 categories and Gemma 3 27B in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Claude Opus 4.6 leads 57.7 to 16.3.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 94.4% for Claude Opus 4.6 and 22.5% for Gemma 3 27B.
  • Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $5 / $25 for Claude Opus 4.6.
  • Claude Opus 4.6 accepts more context: 1M tokens versus 131K.
  • Gemma 3 27B has downloadable open weights; the other is API-only.

Side by side

Claude Opus 4.6Gemma 3 27B
ProviderAnthropicGoogle
Noometry Index58.330.5
Released2026-02-042025-03-11
WeightsProprietaryOpen
Context window1M131K
Max output128K8K
Input $ / M tokens$5$0.08
Output $ / M tokens$25$0.16
Results tracked6843

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

Category by category

Coding Claude Opus 4.6 leads

Claude Opus 4.6: 57.7 (#15), Gemma 3 27B: 22.1 (#329)

BenchmarkClaude Opus 4.6Gemma 3 27B
LMArena Coding15351322
SWE-bench Verified78.7%—
FrontierCode26.6%—
SWE-bench Verified (bash only)75.6%—
Aider Polyglot—4.9%
LMArena WebDev1546—
SWE-bench Multilingual72%—
SciCode—21.2%
GSO41.2%—
WeirdML78%—
LiveBench Coding—39.9%
ALE-Bench996.5—
AlgoTune1.47—

Agentic & Tool Use Claude Opus 4.6 leads

Claude Opus 4.6: 51.0 (#4), Gemma 3 27B: 24.9 (#105)

BenchmarkClaude Opus 4.6Gemma 3 27B
Terminal-Bench79.8%—
APEX-Agents46.3%—
Berkeley Function Calling Leaderboard—29.5%
Remote Labor Index4.2%—
τ²-bench Banking27.3%—
Cybench93%—
DeepResearch Bench55.3%—
GBAEval44.1%—
LMArena Search1253—
METR Time Horizons78.9%—
Vending-Bench 28,018—

Reasoning Claude Opus 4.6 leads

Claude Opus 4.6: 57.7 (#21), Gemma 3 27B: 16.3 (#297)

BenchmarkClaude Opus 4.6Gemma 3 27B
Kagi LLM Benchmark83.6%40.4%
Chess Puzzles17%0%
LMArena Hard Prompts15271340
DTBench91.2%52.5%
LMCA55.8%12.3%
Epoch Capabilities Index155.36130.04
ARC-AGI-269.2%—
SimpleBench67.6%—
NYT Connections (extended)92.1%—
ARC-AGI-194%—
CritPt—0%
EnigmaEval7.6%—
Thematic Generalization80.6%—
EBR-Bench12.7%—
LiveBench Reasoning—43.8%
Mystery Game Puzzles25%—
LiveBench Data Analysis—51.5%
ForecastBench60—
LiveBench—50%

Math Claude Opus 4.6 leads

Claude Opus 4.6: 62.7 (#25), Gemma 3 27B: 24.4 (#261)

BenchmarkClaude Opus 4.6Gemma 3 27B
OTIS Mock AIME 2024-202594.4%22.5%
LMArena Math15171312
FrontierMath (Tiers 1-3)66%—
FrontierMath Tier 426.8%—
MathArena Final-Answer Competitions78.5%—
ProofBench50%—
LiveBench Math—55.4%
MATH Level 5—74%
FrontierMath (Feb 2025 set)40.7%—
FrontierMath Tier 4 (v1)22.9%—

Knowledge Claude Opus 4.6 leads

Claude Opus 4.6: 61.8 (#23), Gemma 3 27B: 25.4 (#256)

BenchmarkClaude Opus 4.6Gemma 3 27B
GPQA Diamond90.5%47.7%
Vectara Hallucination Rate12.2%7.4%
LMArena Expert15461303
Humanity's Last Exam34.4%—
SimpleQA Verified47%—
Confabulations—40.3%

Multimodal Claude Opus 4.6 leads

Claude Opus 4.6: 37.0 (#67), Gemma 3 27B: 32.0 (#94)

BenchmarkClaude Opus 4.6Gemma 3 27B
LMArena Vision13161165
GeoBench—52%
Furniture Assembly28.3%—
LMArena Document1507—

Multilingual Claude Opus 4.6 leads

Claude Opus 4.6: 58.2 (#4), Gemma 3 27B: 47.0 (#150)

BenchmarkClaude Opus 4.6Gemma 3 27B
LMArena Non-English14901334
LMArena Chinese15511347
LMArena French15131368
LMArena German15041362
LMArena Japanese14841287
LMArena Korean14651308
LMArena Russian14971349
LMArena Spanish15121348

Instruction Following Claude Opus 4.6 leads

Claude Opus 4.6: 79.7 (#3), Gemma 3 27B: 70.8 (#155)

BenchmarkClaude Opus 4.6Gemma 3 27B
LMArena Instruction Following15231321
LiveBench Instruction Following—74.9%

Long Context Claude Opus 4.6 leads

Claude Opus 4.6: 48.1 (#13), Gemma 3 27B: 27.6 (#288)

BenchmarkClaude Opus 4.6Gemma 3 27B
LMArena Longer Query15201333
Fiction.LiveBench—33.3%
CL-bench20.7%—
CL-bench Life17%—

Writing & Preference Claude Opus 4.6 leads

Claude Opus 4.6: 73.7 (#9), Gemma 3 27B: 52.7 (#161)

BenchmarkClaude Opus 4.6Gemma 3 27B
LMArena Text15041358
LMArena Creative Writing15061346
EQ-Bench Creative Writing18091266
LMArena Multi-Turn15141344
Short-Story Creative Writing—79.9%
EQ-Bench 41223—
LiveBench Language—34.6%

Frequently asked questions

Is Claude Opus 4.6 better than Gemma 3 27B?

Claude Opus 4.6 is the stronger model overall, scoring 58.3 to 30.5 on the Noometry Index. Gemma 3 27B costs 100× less per token, which makes it the better buy when Claude Opus 4.6's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 4.6 or Gemma 3 27B?

Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Claude Opus 4.6 lists at $5 and $25.

Is Claude Opus 4.6 or Gemma 3 27B better for coding?

Claude Opus 4.6 scores higher on coding benchmarks: 57.7 versus 22.1 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 4.6 does, with 1M tokens against 131K.

How many benchmarks do Claude Opus 4.6 and Gemma 3 27B share?

27 benchmarks have published results for both models. Claude Opus 4.6 has 68 scored results on Noometry and Gemma 3 27B has 43.

Related comparisons

Go deeper