Noometry

Model comparison

Claude Sonnet 4 vs Llama 3.2 90B

Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 27.5 on the Noometry Index.

Last verified . 7 shared benchmarks.

Claude Sonnet 4 Anthropic

40.8

Rank #145 Confirmed

Llama 3.2 90B Meta

27.5

Rank #331 Confirmed

Summary

  • They share 7 benchmarks with published results for both. Claude Sonnet 4 scores higher in 5 categories and Llama 3.2 90B in 0 categories; 4 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Sonnet 4 leads 43.3 to 11.1.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for Claude Sonnet 4 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 Sonnet 4Llama 3.2 90B
ProviderAnthropicMeta
Noometry Index40.827.5
Released2025-05-222024-09-24
WeightsProprietaryOpen
Context window200K—
Max output64K—
Input $ / M tokens$3—
Output $ / M tokens$15—
Results tracked589

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

Category by category

Coding Not comparable

Claude Sonnet 4: 43.5 (#88), Llama 3.2 90B: —

BenchmarkClaude Sonnet 4Llama 3.2 90B
SWE-bench Verified (bash only)64.9%—
Aider Polyglot61.3%—
SciCode40%—
GSO4.9%—
WeirdML46.1%—
LMArena Coding1414—
ALE-Bench655.35—

Agentic & Tool Use Claude Sonnet 4 leads

Claude Sonnet 4: 38.5 (#31), Llama 3.2 90B: 30.0 (#80)

BenchmarkClaude Sonnet 4Llama 3.2 90B
TheAgentCompany33.1%—
Cybench35%—
DeepResearch Bench46.6%—
OSWorld43.9%—
BALROG—27.3%
METR Time Horizons62%—

Reasoning Claude Sonnet 4 leads

Claude Sonnet 4: 22.9 (#187), Llama 3.2 90B: 21.7 (#217)

BenchmarkClaude Sonnet 4Llama 3.2 90B
EnigmaEval3.1%0.4%
Epoch Capabilities Index141.69125.5
ARC-AGI-25.9%—
SimpleBench45.5%—
Kagi LLM Benchmark73%—
ARC-AGI-140%—
CritPt0.3%—
LMArena Hard Prompts1372—
DTBench77.1%—
LMCA29%—
ForecastBench60.2—

Math Claude Sonnet 4 leads

Claude Sonnet 4: 43.3 (#80), Llama 3.2 90B: 11.1 (#308)

BenchmarkClaude Sonnet 4Llama 3.2 90B
OTIS Mock AIME 2024-202571.1%2.6%
MATH Level 584.4%39.4%
Omni-MATH60.2%—
LMArena Math1375—
FrontierMath (Feb 2025 set)4.1%—
FrontierMath Tier 4 (v1)0%—

Knowledge Claude Sonnet 4 leads

Claude Sonnet 4: 41.8 (#108), Llama 3.2 90B: 21.7 (#274)

BenchmarkClaude Sonnet 4Llama 3.2 90B
GPQA Diamond79.2%41%
Humanity's Last Exam7.8%—
MMLU-Pro84.3%—
Confabulations13.2%—
Vectara Hallucination Rate10.3%—
GPQA (HELM)70.6%—
LMArena Expert1372—
MMLU—80.3%

Multimodal Too close to call

Claude Sonnet 4: 26.2 (#121), Llama 3.2 90B: 25.4 (#124)

BenchmarkClaude Sonnet 4Llama 3.2 90B
LMArena Vision11911000
GeoBench37%52%
VPCT34%—
MindCube44.8%—

Multilingual Not comparable

Claude Sonnet 4: 46.7 (#156), Llama 3.2 90B: —

BenchmarkClaude Sonnet 4Llama 3.2 90B
LMArena Non-English1333—
LMArena Chinese1350—
LMArena French1363—
LMArena German1331—
LMArena Japanese1302—
LMArena Korean1291—
LMArena Russian1355—
LMArena Spanish1357—

Instruction Following Not comparable

Claude Sonnet 4: 71.7 (#145), Llama 3.2 90B: —

BenchmarkClaude Sonnet 4Llama 3.2 90B
IFEval84%—
LMArena Instruction Following1376—

Long Context Not comparable

Claude Sonnet 4: 33.7 (#259), Llama 3.2 90B: —

BenchmarkClaude Sonnet 4Llama 3.2 90B
Fiction.LiveBench46.9%—
LMArena Longer Query1398—

Writing & Preference Not comparable

Claude Sonnet 4: 57.1 (#132), Llama 3.2 90B: —

BenchmarkClaude Sonnet 4Llama 3.2 90B
LMArena Text1351—
LMArena Creative Writing1345—
Short-Story Creative Writing81.4%—
EQ-Bench Creative Writing1483—
WildBench83.8%—
LMArena Multi-Turn1376—

Frequently asked questions

Is Claude Sonnet 4 better than Llama 3.2 90B?

Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 27.5 on the Noometry Index.

How many benchmarks do Claude Sonnet 4 and Llama 3.2 90B share?

7 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Llama 3.2 90B has 9.

Related comparisons

Go deeper