Noometry

Model comparison

Claude Sonnet 4.5 vs Grok 4.20 (Non-Reasoning)

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 44.0 on the Noometry Index.

Last verified . 41 shared benchmarks.

Claude Sonnet 4.5 Anthropic

44.0

Rank #76 Confirmed

Grok 4.20 (Non-Reasoning) xAI

48.5

Rank #48 Confirmed

Summary

  • They share 41 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 5 categories and Grok 4.20 (Non-Reasoning) in 5 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.2 to 26.9.
  • The biggest single-benchmark swing is ARC-AGI-2: 13.6% for Claude Sonnet 4.5 and 65.1% for Grok 4.20 (Non-Reasoning).
  • Grok 4.20 (Non-Reasoning) is cheaper at $1.25 / $2.50 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.

Side by side

Claude Sonnet 4.5Grok 4.20 (Non-Reasoning)
ProviderAnthropicxAI
Noometry Index44.048.5
Released2025-09-292026-02-17
WeightsProprietaryProprietary
Context window1M1M
Max output64K30K
Input $ / M tokens$3$1.25
Output $ / M tokens$15$2.50
Results tracked7346

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), Grok 4.20 (Non-Reasoning): 41.8 (#104)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
LMArena WebDev13931374
WeirdML47.7%52.3%
LMArena Coding14891458
ALE-Bench796.151,150
SWE-bench Verified71.3%—
SWE-bench Verified (bash only)71.4%—
SWE-bench Multilingual67%—
SciCode44.7%—
GSO14.7%—
AlgoTune1.52—

Agentic & Tool Use Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 38.1 (#30), Grok 4.20 (Non-Reasoning): 34.3 (#43)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
Terminal-Bench46.5%57.3%
τ²-bench Banking25.3%18%
LMArena Search11591189
Vending-Bench 23,8394,663
Berkeley Function Calling Leaderboard73.2%—
GDPval42.5%—
Remote Labor Index2.1%—
τ²-bench Airline72%—
τ²-bench Retail72.4%—
τ²-bench Telecom84.9%—
Cybench60%—
DeepResearch Bench52.6%—
OSWorld62.9%—
METR Time Horizons67.4%—

Reasoning Grok 4.20 (Non-Reasoning) leads

Claude Sonnet 4.5: 26.9 (#113), Grok 4.20 (Non-Reasoning): 52.2 (#30)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
ARC-AGI-213.6%65.1%
Kagi LLM Benchmark57.9%75%
NYT Connections (extended)37.3%85.4%
ARC-AGI-163.7%89.5%
Chess Puzzles12%24%
LMArena Hard Prompts14621451
DTBench83.2%90.1%
LMCA38.8%38.7%
Epoch Capabilities Index146.84152.01
ForecastBench61.961.4
SimpleBench54.3%—
CritPt1.1%—
EnigmaEval6%—
Thematic Generalization—63.8%
EBR-Bench2.4%—
Mystery Game Puzzles17%—

Math Grok 4.20 (Non-Reasoning) leads

Claude Sonnet 4.5: 31.9 (#190), Grok 4.20 (Non-Reasoning): 48.0 (#54)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
FrontierMath (Tiers 1-3)23.9%44.9%
FrontierMath Tier 42.4%17.1%
OTIS Mock AIME 2024-202577.8%92.2%
ProofBench19%14%
LMArena Math14481455
Omni-MATH55.3%—
MATH Level 597.7%—
FrontierMath (Feb 2025 set)15.2%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge Grok 4.20 (Non-Reasoning) leads

Claude Sonnet 4.5: 48.4 (#70), Grok 4.20 (Non-Reasoning): 52.8 (#54)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
GPQA Diamond82.3%89.3%
SimpleQA Verified30.7%30.2%
LMArena Expert14831440
Humanity's Last Exam13.7%—
MMLU-Pro86.9%—
Vectara Hallucination Rate12%—
GPQA (HELM)68.6%—

Multimodal Too close to call

Claude Sonnet 4.5: 34.1 (#84), Grok 4.20 (Non-Reasoning): 33.2 (#88)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
LMArena Document14501416
LMArena Vision—1263
VPCT39.8%—
Blueprint-Bench 2—0%

Multilingual Grok 4.20 (Non-Reasoning) leads

Claude Sonnet 4.5: 53.6 (#65), Grok 4.20 (Non-Reasoning): 54.7 (#38)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
LMArena Non-English14251440
LMArena Chinese14591481
LMArena French14591477
LMArena German14271463
LMArena Japanese13901450
LMArena Korean14031417
LMArena Russian14361457
LMArena Spanish14571442

Instruction Following Too close to call

Claude Sonnet 4.5: 75.2 (#74), Grok 4.20 (Non-Reasoning): 75.0 (#79)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
LMArena Instruction Following14591420
IFEval85%—

Long Context Too close to call

Claude Sonnet 4.5: 45.2 (#45), Grok 4.20 (Non-Reasoning): 45.4 (#33)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
LMArena Longer Query14761436
CL-bench—22.2%
CL-bench Life—11.9%

Writing & Preference Too close to call

Claude Sonnet 4.5: 66.7 (#33), Grok 4.20 (Non-Reasoning): 65.9 (#42)

BenchmarkClaude Sonnet 4.5Grok 4.20 (Non-Reasoning)
LMArena Text14391451
LMArena Creative Writing14421439
EQ-Bench Creative Writing16781574
LMArena Multi-Turn14651456
WildBench85.4%—

Frequently asked questions

Is Claude Sonnet 4.5 better than Grok 4.20 (Non-Reasoning)?

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 44.0 on the Noometry Index.

Which is cheaper, Claude Sonnet 4.5 or Grok 4.20 (Non-Reasoning)?

Grok 4.20 (Non-Reasoning) is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

Is Claude Sonnet 4.5 or Grok 4.20 (Non-Reasoning) better for coding?

Claude Sonnet 4.5 scores higher on coding benchmarks: 47.8 versus 41.8 in the Noometry coding category.

Which has the bigger context window?

Both accept 1M tokens.

How many benchmarks do Claude Sonnet 4.5 and Grok 4.20 (Non-Reasoning) share?

41 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.

Related comparisons

Go deeper