Noometry

Model comparison

Claude Opus 4.6 vs Gemini 3.1 Pro Preview

Claude Opus 4.6 is the stronger model overall, scoring 58.2 to 56.7 on the Noometry Index. Gemini 3.1 Pro Preview costs 2.2× less per token, which makes it the better buy when Claude Opus 4.6's lead doesn't matter for your workload.

Last verified . 62 shared benchmarks.

Claude Opus 4.6 Anthropic

58.2

Rank #20 Confirmed

Gemini 3.1 Pro Preview Google

56.7

Rank #23 Confirmed

Summary

  • They share 62 benchmarks with published results for both. Claude Opus 4.6 scores higher in 7 categories and Gemini 3.1 Pro Preview in 3 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in coding, where Claude Opus 4.6 leads 57.2 to 42.5.
  • The biggest single-benchmark swing is GBAEval: 44.1% for Claude Opus 4.6 and 0.8% for Gemini 3.1 Pro Preview.
  • Gemini 3.1 Pro Preview is cheaper at $2 / $12 per million input/output tokens, against $5 / $25 for Claude Opus 4.6.
  • Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 1M.

Side by side

Claude Opus 4.6Gemini 3.1 Pro Preview
ProviderAnthropicGoogle
Noometry Index58.256.7
Released2026-02-042026-02-19
WeightsProprietaryProprietary
Context window1M1.05M
Max output128K66K
Input $ / M tokens$5$2
Output $ / M tokens$25$12
Results tracked6871

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: 57.2 (#20), Gemini 3.1 Pro Preview: 42.5 (#99)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
SWE-bench Verified78.7%75.6%
LMArena WebDev15471447
GSO41.2%22.6%
WeirdML78%72.1%
LMArena Coding15361484
ALE-Bench996.51,161
AlgoTune1.472.02
DeepSWE—11.7%
FrontierCode26.6%—
SWE-bench Verified (bash only)75.6%—
SWE-bench Multilingual72%—
SciCode—58.9%
MirrorCode—8.9%

Agentic & Tool Use Claude Opus 4.6 leads

Claude Opus 4.6: 51.1 (#4), Gemini 3.1 Pro Preview: 37.7 (#34)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
Terminal-Bench79.8%80.2%
APEX-Agents46.3%35.3%
τ²-bench Banking27.3%26%
DeepResearch Bench55.3%47.8%
GBAEval44.1%0.8%
LMArena Search12531211
METR Time Horizons78.9%77%
Vending-Bench 28,0183,774
Remote Labor Index4.2%—
Cybench93%—
PostTrainBench—22%
BALROG—57%
ExploitBench—26.1%
GDP.pdf—17%

Reasoning Gemini 3.1 Pro Preview leads

Claude Opus 4.6: 57.8 (#23), Gemini 3.1 Pro Preview: 71.7 (#12)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
ARC-AGI-269.2%77.1%
SimpleBench67.6%79.6%
NYT Connections (extended)92.1%97.4%
ARC-AGI-194%98%
Chess Puzzles17%55%
EnigmaEval7.6%36.8%
Thematic Generalization80.6%79.4%
EBR-Bench12.7%14.3%
LMArena Hard Prompts15271485
Mystery Game Puzzles25%34%
DTBench91.2%97.1%
LMCA55.8%53.8%
Epoch Capabilities Index155.24154.77
ForecastBench6059
Kagi LLM Benchmark83.6%—
CritPt—17.7%

Math Too close to call

Claude Opus 4.6: 63.0 (#31), Gemini 3.1 Pro Preview: 62.1 (#34)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
FrontierMath (Tiers 1-3)66%59.6%
FrontierMath Tier 426.8%26.8%
MathArena Final-Answer Competitions78.5%86.5%
OTIS Mock AIME 2024-202594.4%95.6%
ProofBench50%26%
LMArena Math15191485
FrontierMath (Feb 2025 set)40.7%36.9%
FrontierMath Tier 4 (v1)22.9%16.7%

Knowledge Gemini 3.1 Pro Preview leads

Claude Opus 4.6: 61.9 (#26), Gemini 3.1 Pro Preview: 71.8 (#3)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
GPQA Diamond90.5%94.4%
Humanity's Last Exam34.4%46.4%
SimpleQA Verified47%73.5%
Vectara Hallucination Rate12.2%10.4%
LMArena Expert15461485

Multimodal Too close to call

Claude Opus 4.6: 37.3 (#74), Gemini 3.1 Pro Preview: 37.9 (#69)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
LMArena Vision13161296
Furniture Assembly28.3%26.7%
LMArena Document15071444
Blueprint-Bench 2—26.5%

Multilingual Too close to call

Claude Opus 4.6: 57.9 (#6), Gemini 3.1 Pro Preview: 57.0 (#12)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
LMArena Non-English14891477
LMArena Chinese15511529
LMArena French15131487
LMArena German15021491
LMArena Japanese14841493
LMArena Korean14641455
LMArena Russian14971498
LMArena Spanish15101479

Instruction Following Claude Opus 4.6 leads

Claude Opus 4.6: 79.5 (#4), Gemini 3.1 Pro Preview: 77.0 (#32)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
LMArena Instruction Following15231466

Long Context Too close to call

Claude Opus 4.6: 48.1 (#13), Gemini 3.1 Pro Preview: 47.4 (#18)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
CL-bench20.7%20.8%
CL-bench Life17%16.9%
LMArena Longer Query15201483

Writing & Preference Claude Opus 4.6 leads

Claude Opus 4.6: 73.5 (#10), Gemini 3.1 Pro Preview: 66.1 (#37)

BenchmarkClaude Opus 4.6Gemini 3.1 Pro Preview
LMArena Text15031481
LMArena Creative Writing15051482
EQ-Bench Creative Writing18091491
EQ-Bench 412231142
LMArena Multi-Turn15131488

Frequently asked questions

Is Claude Opus 4.6 better than Gemini 3.1 Pro Preview?

Claude Opus 4.6 is the stronger model overall, scoring 58.2 to 56.7 on the Noometry Index. Gemini 3.1 Pro Preview costs 2.2× 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 Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview is cheaper. It lists at $2 per million input tokens and $12 per million output tokens; Claude Opus 4.6 lists at $5 and $25.

Is Claude Opus 4.6 or Gemini 3.1 Pro Preview better for coding?

Claude Opus 4.6 scores higher on coding benchmarks: 57.2 versus 42.5 in the Noometry coding category.

Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 1M.

How many benchmarks do Claude Opus 4.6 and Gemini 3.1 Pro Preview share?

62 benchmarks have published results for both models. Claude Opus 4.6 has 68 scored results on Noometry and Gemini 3.1 Pro Preview has 71.

Related comparisons

Go deeper