Noometry

Model comparison

Claude Opus 4.5 vs Grok 4.20 (Non-Reasoning)

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 48.6 on the Noometry Index. Grok 4.20 (Non-Reasoning) costs 6.4× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

Last verified . 42 shared benchmarks.

Claude Opus 4.5 Anthropic

50.5

Rank #47 Confirmed

Grok 4.20 (Non-Reasoning) xAI

48.6

Rank #54 Confirmed

Summary

  • They share 42 benchmarks with published results for both. Claude Opus 4.5 scores higher in 6 categories and Grok 4.20 (Non-Reasoning) in 4 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in agentic & tool use, where Claude Opus 4.5 leads 47.3 to 34.4.
  • The biggest single-benchmark swing is NYT Connections (extended): 52.5% for Claude Opus 4.5 and 85.4% for Grok 4.20 (Non-Reasoning).
  • Grok 4.20 (Non-Reasoning) is cheaper at $1.25 / $2.50 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
  • Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 200K.

Side by side

Claude Opus 4.5Grok 4.20 (Non-Reasoning)
ProviderAnthropicxAI
Noometry Index50.548.6
Released2025-11-012026-02-17
WeightsProprietaryProprietary
Context window200K1M
Max output64K30K
Input $ / M tokens$5$1.25
Output $ / M tokens$25$2.50
Results tracked6946

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

Category by category

Coding Claude Opus 4.5 leads

Claude Opus 4.5: 54.8 (#27), Grok 4.20 (Non-Reasoning): 42.1 (#112)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
LMArena WebDev14941375
WeirdML63.7%52.3%
LMArena Coding15041459
ALE-Bench1,0251,150
SWE-bench Verified76.7%—
SWE-bench Verified (bash only)76.8%—
SWE-bench Multilingual70.7%—
GSO26.5%—
AlgoTune1.77—

Agentic & Tool Use Claude Opus 4.5 leads

Claude Opus 4.5: 47.3 (#12), Grok 4.20 (Non-Reasoning): 34.4 (#46)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
Terminal-Bench63.1%57.3%
τ²-bench Banking24.7%18%
LMArena Search11801189
Vending-Bench 24,9674,663
Berkeley Function Calling Leaderboard77.5%—
GDPval45.5%—
Remote Labor Index3.8%—
τ²-bench Airline84%—
τ²-bench Retail79.6%—
τ²-bench Telecom92.3%—
Cybench82%—
DeepResearch Bench54.8%—
OSWorld66.3%—
BALROG43.5%—
METR Time Horizons75%—

Reasoning Grok 4.20 (Non-Reasoning) leads

Claude Opus 4.5: 42.6 (#51), Grok 4.20 (Non-Reasoning): 52.3 (#32)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
ARC-AGI-237.6%65.1%
Kagi LLM Benchmark80.2%75%
NYT Connections (extended)52.5%85.4%
ARC-AGI-180%89.5%
Chess Puzzles12%24%
LMArena Hard Prompts14761451
DTBench89.9%90.1%
LMCA44.5%38.7%
Epoch Capabilities Index150.09151.98
ForecastBench60.761.4
SimpleBench62%—
EnigmaEval11.9%—
Thematic Generalization—63.8%
EBR-Bench14.3%—
Mystery Game Puzzles22%—

Math Grok 4.20 (Non-Reasoning) leads

Claude Opus 4.5: 38.6 (#132), Grok 4.20 (Non-Reasoning): 48.2 (#65)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
FrontierMath (Tiers 1-3)34.4%44.9%
FrontierMath Tier 44.9%17.1%
OTIS Mock AIME 2024-202586.1%92.2%
ProofBench36%14%
LMArena Math14631455
FrontierMath (Feb 2025 set)20.7%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge Claude Opus 4.5 leads

Claude Opus 4.5: 56.5 (#44), Grok 4.20 (Non-Reasoning): 52.8 (#60)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
GPQA Diamond86%89.3%
SimpleQA Verified45.7%30.2%
LMArena Expert14871439
Humanity's Last Exam25.2%—
Vectara Hallucination Rate10.9%—

Multimodal Grok 4.20 (Non-Reasoning) leads

Claude Opus 4.5: 31.4 (#107), Grok 4.20 (Non-Reasoning): 33.3 (#98)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
LMArena Document14621416
LMArena Vision—1263
GeoBench75%—
VPCT40%—
Blueprint-Bench 2—0%
Furniture Assembly28.3%—

Multilingual Too close to call

Claude Opus 4.5: 54.3 (#47), Grok 4.20 (Non-Reasoning): 54.5 (#40)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
LMArena Non-English14381441
LMArena Chinese14701481
LMArena French14711476
LMArena German14491465
LMArena Japanese14161449
LMArena Korean14241417
LMArena Russian14471458
LMArena Spanish14581443

Instruction Following Claude Opus 4.5 leads

Claude Opus 4.5: 77.5 (#19), Grok 4.20 (Non-Reasoning): 74.8 (#83)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
LMArena Instruction Following14781420

Long Context Too close to call

Claude Opus 4.5: 46.5 (#22), Grok 4.20 (Non-Reasoning): 45.5 (#34)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
CL-bench21.1%22.2%
LMArena Longer Query14801437
CL-bench Life—11.9%

Writing & Preference Claude Opus 4.5 leads

Claude Opus 4.5: 68.1 (#28), Grok 4.20 (Non-Reasoning): 65.7 (#44)

BenchmarkClaude Opus 4.5Grok 4.20 (Non-Reasoning)
LMArena Text14511451
LMArena Creative Writing14451438
EQ-Bench Creative Writing16871574
LMArena Multi-Turn14661456

Frequently asked questions

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

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 48.6 on the Noometry Index. Grok 4.20 (Non-Reasoning) costs 6.4× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 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 Opus 4.5 lists at $5 and $25.

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

Claude Opus 4.5 scores higher on coding benchmarks: 54.8 versus 42.1 in the Noometry coding category.

Which has the bigger context window?

Grok 4.20 (Non-Reasoning) does, with 1M tokens against 200K.

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

42 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.

Related comparisons

Go deeper