Noometry

Model comparison

Claude Sonnet 4.5 vs DeepSeek-V3.2-Exp

Claude Sonnet 4.5 and DeepSeek-V3.2-Exp score almost the same on the Noometry Index (44.1 vs 44.3), so choose on price, context window or the category you care about most.

Last verified . 41 shared benchmarks.

Claude Sonnet 4.5 Anthropic

44.1

Rank #81 Confirmed

DeepSeek-V3.2-Exp DeepSeek

44.3

Rank #78 Confirmed

Summary

  • They share 41 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 6 categories and DeepSeek-V3.2-Exp in 3 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where DeepSeek-V3.2-Exp leads 41.7 to 32.3.
  • The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 73.2% for Claude Sonnet 4.5 and 56.7% for DeepSeek-V3.2-Exp.
  • DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
  • Claude Sonnet 4.5 accepts more context: 200K tokens versus 164K.
  • DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.

Side by side

Claude Sonnet 4.5DeepSeek-V3.2-Exp
ProviderAnthropicDeepSeek
Noometry Index44.144.3
Released2025-09-292025-09-29
WeightsProprietaryOpen
Context window200K164K
Max output64K66K
Input $ / M tokens$3$0.26
Output $ / M tokens$15$0.38
Results tracked7349

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

Category by category

Coding Too close to call

Claude Sonnet 4.5: 47.3 (#61), DeepSeek-V3.2-Exp: 46.5 (#65)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
SWE-bench Verified (bash only)71.4%70%
LMArena WebDev13931362
SWE-bench Multilingual67%59%
SciCode44.7%38.9%
WeirdML47.7%39.5%
LMArena Coding14891454
SWE-bench Verified71.3%—
Aider Polyglot—74.2%
GSO14.7%—
ALE-Bench796.15—
AlgoTune1.52—

Agentic & Tool Use Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 38.3 (#32), DeepSeek-V3.2-Exp: 32.7 (#59)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
Terminal-Bench46.5%39.6%
Berkeley Function Calling Leaderboard73.2%56.7%
Vending-Bench 23,8391,034
APEX-Agents—21.3%
GDPval42.5%—
Remote Labor Index2.1%—
TheAgentCompany—42.9%
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%—

Reasoning Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 26.9 (#125), DeepSeek-V3.2-Exp: 22.1 (#208)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
ARC-AGI-213.6%4%
Kagi LLM Benchmark57.9%52.2%
NYT Connections (extended)37.3%36.7%
ARC-AGI-163.7%57%
CritPt1.1%2.9%
Chess Puzzles12%14%
LMArena Hard Prompts14621434
DTBench83.2%87.7%
LMCA38.8%29.1%
Epoch Capabilities Index146.84146.27
SimpleBench54.3%—
EnigmaEval6%—
Thematic Generalization—65%
EBR-Bench2.4%—
Mystery Game Puzzles17%—
ForecastBench61.9—

Math DeepSeek-V3.2-Exp leads

Claude Sonnet 4.5: 32.3 (#216), DeepSeek-V3.2-Exp: 41.7 (#87)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
OTIS Mock AIME 2024-202577.8%87.8%
ProofBench19%8%
LMArena Math14491435
FrontierMath (Feb 2025 set)15.2%22.1%
FrontierMath Tier 4 (v1)4.2%2.1%
FrontierMath (Tiers 1-3)23.9%—
FrontierMath Tier 42.4%—
MathArena Final-Answer Competitions—57.7%
Omni-MATH55.3%—
MATH Level 597.7%—

Knowledge DeepSeek-V3.2-Exp leads

Claude Sonnet 4.5: 48.4 (#76), DeepSeek-V3.2-Exp: 51.7 (#66)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
GPQA Diamond82.3%83.4%
Vectara Hallucination Rate12%5.3%
LMArena Expert14821436
Humanity's Last Exam13.7%—
SimpleQA Verified30.7%—
MMLU-Pro86.9%—
GPQA (HELM)68.6%—

Multimodal Not comparable

Claude Sonnet 4.5: 34.8 (#89), DeepSeek-V3.2-Exp: —

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
VPCT39.8%—
LMArena Document1450—

Multilingual Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 53.4 (#69), DeepSeek-V3.2-Exp: 52.2 (#90)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
LMArena Non-English14251409
LMArena Chinese14591461
LMArena French14581433
LMArena German14271440
LMArena Japanese13901374
LMArena Korean14031371
LMArena Russian14371424
LMArena Spanish14571440

Instruction Following Too close to call

Claude Sonnet 4.5: 75.0 (#78), DeepSeek-V3.2-Exp: 74.5 (#93)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
LMArena Instruction Following14591413
IFEval85%—

Long Context DeepSeek-V3.2-Exp leads

Claude Sonnet 4.5: 45.2 (#46), DeepSeek-V3.2-Exp: 47.6 (#16)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
LMArena Longer Query14761428
Fiction.LiveBench—83.3%
CL-bench—13.2%
CL-bench Life—9.5%

Writing & Preference Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 66.5 (#34), DeepSeek-V3.2-Exp: 62.4 (#77)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.2-Exp
LMArena Text14391425
LMArena Creative Writing14421403
EQ-Bench Creative Writing16781515
LMArena Multi-Turn14651427
WildBench85.4%—

Frequently asked questions

Is Claude Sonnet 4.5 better than DeepSeek-V3.2-Exp?

Claude Sonnet 4.5 and DeepSeek-V3.2-Exp score almost the same on the Noometry Index (44.1 vs 44.3), so choose on price, context window or the category you care about most.

Which is cheaper, Claude Sonnet 4.5 or DeepSeek-V3.2-Exp?

DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

Is Claude Sonnet 4.5 or DeepSeek-V3.2-Exp better for coding?

They score almost the same on coding (47.3 vs 46.5); test both on your own repository before choosing.

Which has the bigger context window?

Claude Sonnet 4.5 does, with 200K tokens against 164K.

How many benchmarks do Claude Sonnet 4.5 and DeepSeek-V3.2-Exp share?

41 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and DeepSeek-V3.2-Exp has 49.

Related comparisons

Go deeper