Noometry

Model comparison

Claude Sonnet 4.5 vs Llama 4 Maverick

Claude Sonnet 4.5 is the stronger model overall, scoring 44.0 to 30.6 on the Noometry Index. Llama 4 Maverick costs 20× less per token, which makes it the better buy when Claude Sonnet 4.5's lead doesn't matter for your workload.

Last verified . 45 shared benchmarks.

Claude Sonnet 4.5 Anthropic

44.0

Rank #76 Confirmed

Llama 4 Maverick Meta

30.6

Rank #276 Confirmed

Summary

  • They share 45 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 10 categories and Llama 4 Maverick in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Claude Sonnet 4.5 leads 66.7 to 39.0.
  • The biggest single-benchmark swing is ARC-AGI-1: 63.7% for Claude Sonnet 4.5 and 4.4% for Llama 4 Maverick.
  • Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
  • Claude Sonnet 4.5 accepts more context: 1M tokens versus 128K.
  • Llama 4 Maverick has downloadable open weights; the other is API-only.

Side by side

Claude Sonnet 4.5Llama 4 Maverick
ProviderAnthropicMeta
Noometry Index44.030.6
Released2025-09-292025-04-05
WeightsProprietaryOpen
Context window1M128K
Max output64K4K
Input $ / M tokens$3$0.19
Output $ / M tokens$15$0.65
Results tracked7354

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

Category by category

Coding Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 47.8 (#54), Llama 4 Maverick: 26.3 (#318)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
SWE-bench Verified (bash only)71.4%21%
SciCode44.7%33.1%
WeirdML47.7%24.5%
LMArena Coding14891302
ALE-Bench796.15172.97
SWE-bench Verified71.3%—
Aider Polyglot—15.6%
LMArena WebDev1393—
SWE-bench Multilingual67%—
GSO14.7%—
BigCodeBench Instruct—49.7%
BigCodeBench Complete—61.4%
AlgoTune1.52—

Agentic & Tool Use Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 38.1 (#30), Llama 4 Maverick: 27.9 (#86)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
Berkeley Function Calling Leaderboard73.2%37.3%
Terminal-Bench46.5%—
GDPval42.5%—
Remote Labor Index2.1%—
τ²-bench Airline72%—
τ²-bench Banking25.3%—
τ²-bench Retail72.4%—
τ²-bench Telecom84.9%—
Cybench60%—
DeepResearch Bench52.6%—
OSWorld62.9%—
LMArena Search1159—
METR Time Horizons67.4%—
Vending-Bench 23,839—

Reasoning Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 26.9 (#113), Llama 4 Maverick: 10.1 (#337)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
ARC-AGI-213.6%0%
SimpleBench54.3%27.7%
Kagi LLM Benchmark57.9%55.9%
NYT Connections (extended)37.3%8%
ARC-AGI-163.7%4.4%
CritPt1.1%0%
EnigmaEval6%0.6%
LMArena Hard Prompts14621281
DTBench83.2%61.9%
LMCA38.8%15.9%
Epoch Capabilities Index146.84132.2
ForecastBench61.957.5
Chess Puzzles12%—
EBR-Bench2.4%—
Mystery Game Puzzles17%—

Math Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 31.9 (#190), Llama 4 Maverick: 24.6 (#256)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
OTIS Mock AIME 2024-202577.8%20.6%
Omni-MATH55.3%42.2%
LMArena Math14481299
MATH Level 597.7%73%
FrontierMath (Feb 2025 set)15.2%0.7%
FrontierMath (Tiers 1-3)23.9%—
FrontierMath Tier 42.4%—
ProofBench19%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 48.4 (#70), Llama 4 Maverick: 33.3 (#196)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
GPQA Diamond82.3%67%
Humanity's Last Exam13.7%5.7%
MMLU-Pro86.9%81%
Vectara Hallucination Rate12%8.2%
GPQA (HELM)68.6%65%
LMArena Expert14831259
SimpleQA Verified30.7%—
Confabulations—22.6%

Multimodal Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 34.1 (#84), Llama 4 Maverick: 30.9 (#100)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
LMArena Vision—1141
GeoBench—52%
VPCT39.8%—
LMArena Document1450—
SpatialViz-Bench—31.8%

Multilingual Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 53.6 (#65), Llama 4 Maverick: 42.4 (#190)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
LMArena Non-English14251269
LMArena Chinese14591277
LMArena French14591258
LMArena German14271291
LMArena Japanese13901207
LMArena Korean14031203
LMArena Russian14361286
LMArena Spanish14571293

Instruction Following Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 75.2 (#74), Llama 4 Maverick: 71.9 (#140)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
IFEval85%90.8%
LMArena Instruction Following14591267

Long Context Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 45.2 (#45), Llama 4 Maverick: 31.4 (#274)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
LMArena Longer Query14761280
Fiction.LiveBench—46.2%

Writing & Preference Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 66.7 (#33), Llama 4 Maverick: 39.0 (#247)

BenchmarkClaude Sonnet 4.5Llama 4 Maverick
LMArena Text14391287
LMArena Creative Writing14421266
EQ-Bench Creative Writing1678860
WildBench85.4%80%
LMArena Multi-Turn14651289
Short-Story Creative Writing—62%

Frequently asked questions

Is Claude Sonnet 4.5 better than Llama 4 Maverick?

Claude Sonnet 4.5 is the stronger model overall, scoring 44.0 to 30.6 on the Noometry Index. Llama 4 Maverick costs 20× less per token, which makes it the better buy when Claude Sonnet 4.5's lead doesn't matter for your workload.

Which is cheaper, Claude Sonnet 4.5 or Llama 4 Maverick?

Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

Is Claude Sonnet 4.5 or Llama 4 Maverick better for coding?

Claude Sonnet 4.5 scores higher on coding benchmarks: 47.8 versus 26.3 in the Noometry coding category.

Which has the bigger context window?

Claude Sonnet 4.5 does, with 1M tokens against 128K.

How many benchmarks do Claude Sonnet 4.5 and Llama 4 Maverick share?

45 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and Llama 4 Maverick has 54.

Related comparisons

Go deeper