Noometry

Model comparison

Claude Opus 4.5 vs DeepSeek V4 Pro

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 50.5 on the Noometry Index.

Last verified . 40 shared benchmarks.

Claude Opus 4.5 Anthropic

50.5

Rank #47 Confirmed

DeepSeek V4 Pro DeepSeek

54.3

Rank #31 Confirmed

Summary

  • They share 40 benchmarks with published results for both. Claude Opus 4.5 scores higher in 5 categories and DeepSeek V4 Pro in 4 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 38.6.
  • The biggest single-benchmark swing is NYT Connections (extended): 52.5% for Claude Opus 4.5 and 91.3% for DeepSeek V4 Pro.
  • DeepSeek V4 Pro is cheaper at $0.66 / $1.98 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
  • DeepSeek V4 Pro accepts more context: 1M tokens versus 200K.
  • DeepSeek V4 Pro has downloadable open weights; the other is API-only.

Side by side

Claude Opus 4.5DeepSeek V4 Pro
ProviderAnthropicDeepSeek
Noometry Index50.554.3
Released2025-11-012026-04-24
WeightsProprietaryOpen
Context window200K1M
Max output64K393K
Input $ / M tokens$5$0.66
Output $ / M tokens$25$1.98
Results tracked6948

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), DeepSeek V4 Pro: 52.4 (#34)

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
SWE-bench Verified76.7%77.6%
LMArena WebDev14941582
WeirdML63.7%66.2%
LMArena Coding15041470
ALE-Bench1,0251,403
FrontierCode—28.6%
SWE-bench Verified (bash only)76.8%—
SWE-bench Multilingual70.7%—
SciCode—51%
GSO26.5%—
AlgoTune1.77—

Agentic & Tool Use Claude Opus 4.5 leads

Claude Opus 4.5: 47.3 (#12), DeepSeek V4 Pro: 32.8 (#58)

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
Vending-Bench 24,9673,285
Terminal-Bench63.1%—
APEX-Agents—47.3%
Berkeley Function Calling Leaderboard77.5%—
GDPval45.5%—
Remote Labor Index3.8%—
τ²-bench Airline84%—
τ²-bench Banking24.7%—
τ²-bench Retail79.6%—
τ²-bench Telecom92.3%—
Cybench82%—
DeepResearch Bench54.8%—
OSWorld66.3%—
BALROG43.5%—
LMArena Search1180—
METR Time Horizons75%—

Reasoning DeepSeek V4 Pro leads

Claude Opus 4.5: 42.6 (#51), DeepSeek V4 Pro: 56.5 (#24)

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
ARC-AGI-237.6%61.3%
Kagi LLM Benchmark80.2%53.5%
NYT Connections (extended)52.5%91.3%
ARC-AGI-180%90.5%
Chess Puzzles12%47%
LMArena Hard Prompts14761461
Mystery Game Puzzles22%43%
DTBench89.9%93.9%
LMCA44.5%45.5%
Epoch Capabilities Index150.09155.31
ForecastBench60.756.1
SimpleBench62%—
CritPt—18%
EnigmaEval11.9%—
EBR-Bench14.3%—
Surface Evolver Bench—40%

Math DeepSeek V4 Pro leads

Claude Opus 4.5: 38.6 (#132), DeepSeek V4 Pro: 64.8 (#30)

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
FrontierMath (Tiers 1-3)34.4%64.6%
FrontierMath Tier 44.9%26.8%
OTIS Mock AIME 2024-202586.1%98.6%
ProofBench36%50%
LMArena Math14631455
MathArena Final-Answer Competitions—76.6%
FrontierMath (Feb 2025 set)20.7%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge DeepSeek V4 Pro leads

Claude Opus 4.5: 56.5 (#44), DeepSeek V4 Pro: 59.5 (#31)

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
GPQA Diamond86%91.7%
SimpleQA Verified45.7%52.9%
Vectara Hallucination Rate10.9%8.6%
LMArena Expert14871464
Humanity's Last Exam25.2%—

Multimodal Not comparable

Claude Opus 4.5: 31.4 (#107), DeepSeek V4 Pro: —

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
GeoBench75%—
VPCT40%—
Furniture Assembly28.3%—
LMArena Document1462—

Multilingual Too close to call

Claude Opus 4.5: 54.3 (#47), DeepSeek V4 Pro: 54.4 (#45)

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
LMArena Non-English14381439
LMArena Chinese14701486
LMArena French14711472
LMArena German14491458
LMArena Japanese14161445
LMArena Korean14241447
LMArena Russian14471453
LMArena Spanish14581458

Instruction Following Claude Opus 4.5 leads

Claude Opus 4.5: 77.5 (#19), DeepSeek V4 Pro: 76.1 (#47)

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
LMArena Instruction Following14781448

Long Context Claude Opus 4.5 leads

Claude Opus 4.5: 46.5 (#22), DeepSeek V4 Pro: 45.0 (#51)

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
LMArena Longer Query14801458
CL-bench21.1%—
CL-bench Life—13.5%

Writing & Preference Claude Opus 4.5 leads

Claude Opus 4.5: 68.1 (#28), DeepSeek V4 Pro: 65.5 (#46)

BenchmarkClaude Opus 4.5DeepSeek V4 Pro
LMArena Text14511451
LMArena Creative Writing14451446
EQ-Bench Creative Writing16871553
LMArena Multi-Turn14661467
EQ-Bench 4—1166

Frequently asked questions

Is Claude Opus 4.5 better than DeepSeek V4 Pro?

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 50.5 on the Noometry Index.

Which is cheaper, Claude Opus 4.5 or DeepSeek V4 Pro?

DeepSeek V4 Pro is cheaper. It lists at $0.66 per million input tokens and $1.98 per million output tokens; Claude Opus 4.5 lists at $5 and $25.

Is Claude Opus 4.5 or DeepSeek V4 Pro better for coding?

Claude Opus 4.5 scores higher on coding benchmarks: 54.8 versus 52.4 in the Noometry coding category.

Which has the bigger context window?

DeepSeek V4 Pro does, with 1M tokens against 200K.

How many benchmarks do Claude Opus 4.5 and DeepSeek V4 Pro share?

40 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and DeepSeek V4 Pro has 48.

Related comparisons

Go deeper