Noometry

Model comparison

Claude Opus 5 vs Llama 3.2 90B

Claude Opus 5 is the stronger model overall, scoring 67.7 to 27.1 on the Noometry Index.

Last verified . 5 shared benchmarks.

Claude Opus 5 Anthropic

67.7

Rank #4 Confirmed

Llama 3.2 90B Meta

27.1

Rank #326 Confirmed

Summary

  • They share 5 benchmarks with published results for both. Claude Opus 5 scores higher in 5 categories and Llama 3.2 90B in 0 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Opus 5 leads 86.1 to 10.4.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.9% for Claude Opus 5 and 2.6% for Llama 3.2 90B.
  • Llama 3.2 90B has downloadable open weights; the other is API-only.

Side by side

Claude Opus 5Llama 3.2 90B
ProviderAnthropicMeta
Noometry Index67.727.1
Released2026-07-242024-09-24
WeightsProprietaryOpen
Context window1M—
Max output128K—
Input $ / M tokens$5—
Output $ / M tokens$25—
Results tracked579

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

Category by category

Coding Not comparable

Claude Opus 5: 67.0 (#5), Llama 3.2 90B: —

BenchmarkClaude Opus 5Llama 3.2 90B
DeepSWE73.6%—
FrontierCode53.4%—
CursorBench46.6%—
LMArena WebDev1694—
FrontierSWE52%—
SciCode56.4%—
WeirdML91.8%—
LMArena Coding1533—
ALE-Bench2,165—

Agentic & Tool Use Claude Opus 5 leads

Claude Opus 5: 55.5 (#1), Llama 3.2 90B: 29.7 (#73)

BenchmarkClaude Opus 5Llama 3.2 90B
BALROG63.4%27.3%
APEX-Agents65.8%—
OSWorld 2.031.4%—
τ²-bench Banking48.7%—
PostTrainBench35%—
GBAEval79.6%—
GDP.pdf24%—
Vending-Bench 211,182—

Reasoning Claude Opus 5 leads

Claude Opus 5: 77.1 (#3), Llama 3.2 90B: 20.9 (#207)

BenchmarkClaude Opus 5Llama 3.2 90B
Epoch Capabilities Index162.67125.5
ARC-AGI-290.4%—
SimpleBench80.6%—
NYT Connections (extended)94.3%—
ARC-AGI-197.5%—
CritPt29.1%—
Chess Puzzles42%—
EnigmaEval—0.4%
EBR-Bench45.7%—
LMArena Hard Prompts1525—
Mystery Game Puzzles59%—
DTBench97.9%—
LMCA64.5%—
Bench to the Future 30.12—

Math Claude Opus 5 leads

Claude Opus 5: 86.1 (#6), Llama 3.2 90B: 10.4 (#302)

BenchmarkClaude Opus 5Llama 3.2 90B
OTIS Mock AIME 2024-202598.9%2.6%
FrontierMath (Tiers 1-3)85.6%—
FrontierMath Tier 473.2%—
ProofBench99%—
LMArena Math1533—
MATH Level 5—39.4%

Knowledge Claude Opus 5 leads

Claude Opus 5: 66.8 (#8), Llama 3.2 90B: 21.7 (#269)

BenchmarkClaude Opus 5Llama 3.2 90B
GPQA Diamond93.9%41%
SimpleQA Verified59.9%—
LMArena Expert1558—
MMLU—80.3%

Multimodal Claude Opus 5 leads

Claude Opus 5: 51.1 (#6), Llama 3.2 90B: 24.9 (#119)

BenchmarkClaude Opus 5Llama 3.2 90B
LMArena Vision13211000
GeoBench—52%
Blueprint-Bench 230.4%—
Furniture Assembly60.8%—
LMArena Document1516—

Multilingual Not comparable

Claude Opus 5: 58.8 (#3), Llama 3.2 90B: —

BenchmarkClaude Opus 5Llama 3.2 90B
LMArena Non-English1500—
LMArena Chinese1575—
LMArena French1522—
LMArena German1522—
LMArena Japanese1510—
LMArena Korean1526—
LMArena Russian1507—
LMArena Spanish1519—

Instruction Following Not comparable

Claude Opus 5: 79.3 (#5), Llama 3.2 90B: —

BenchmarkClaude Opus 5Llama 3.2 90B
LMArena Instruction Following1516—

Long Context Not comparable

Claude Opus 5: 46.5 (#20), Llama 3.2 90B: —

BenchmarkClaude Opus 5Llama 3.2 90B
LMArena Longer Query1516—

Writing & Preference Not comparable

Claude Opus 5: 79.3 (#2), Llama 3.2 90B: —

BenchmarkClaude Opus 5Llama 3.2 90B
LMArena Text1506—
LMArena Creative Writing1491—
EQ-Bench Creative Writing2133—
EQ-Bench 41385—
LMArena Multi-Turn1496—

Frequently asked questions

Is Claude Opus 5 better than Llama 3.2 90B?

Claude Opus 5 is the stronger model overall, scoring 67.7 to 27.1 on the Noometry Index.

How many benchmarks do Claude Opus 5 and Llama 3.2 90B share?

5 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Llama 3.2 90B has 9.

Related comparisons

Go deeper