Noometry

Model comparison

Claude Sonnet 4.5 vs DeepSeek-V3.1-Terminus

Claude Sonnet 4.5 and DeepSeek-V3.1-Terminus score almost the same on the Noometry Index (44.1 vs 43.1), so choose on price, context window or the category you care about most.

Last verified . 16 shared benchmarks.

Claude Sonnet 4.5 Anthropic

44.1

Rank #81 Confirmed

DeepSeek-V3.1-Terminus DeepSeek

43.1

Rank #97 Confirmed

Summary

  • They share 16 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 6 categories and DeepSeek-V3.1-Terminus in 1 category; 5 gaps are clear of the uncertainty.
  • The widest gap is in math, where DeepSeek-V3.1-Terminus leads 38.5 to 32.3.
  • The biggest single-benchmark swing is LMCA: 38.8% for Claude Sonnet 4.5 and 28.6% for DeepSeek-V3.1-Terminus.
  • DeepSeek-V3.1-Terminus is cheaper at $0.27 / $1 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-V3.1-Terminus has downloadable open weights; the other is API-only.

Side by side

Claude Sonnet 4.5DeepSeek-V3.1-Terminus
ProviderAnthropicDeepSeek
Noometry Index44.143.1
Released2025-09-292025-09-22
WeightsProprietaryOpen
Context window200K164K
Max output64K147K
Input $ / M tokens$3$0.27
Output $ / M tokens$15$1
Results tracked7316

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: 47.3 (#61), DeepSeek-V3.1-Terminus: 42.0 (#113)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
SciCode44.7%40.6%
LMArena Coding14891426
ALE-Bench796.15745.17
SWE-bench Verified71.3%—
SWE-bench Verified (bash only)71.4%—
LMArena WebDev1393—
SWE-bench Multilingual67%—
GSO14.7%—
WeirdML47.7%—
AlgoTune1.52—

Agentic & Tool Use Not comparable

Claude Sonnet 4.5: 38.3 (#32), DeepSeek-V3.1-Terminus: —

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
Terminal-Bench46.5%—
Berkeley Function Calling Leaderboard73.2%—
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 Too close to call

Claude Sonnet 4.5: 26.9 (#125), DeepSeek-V3.1-Terminus: 26.4 (#133)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
Kagi LLM Benchmark57.9%57.4%
CritPt1.1%1.7%
LMArena Hard Prompts14621426
DTBench83.2%81.3%
LMCA38.8%28.6%
ARC-AGI-213.6%—
SimpleBench54.3%—
NYT Connections (extended)37.3%—
ARC-AGI-163.7%—
Chess Puzzles12%—
EnigmaEval6%—
EBR-Bench2.4%—
Mystery Game Puzzles17%—
Epoch Capabilities Index146.84—
ForecastBench61.9—

Math DeepSeek-V3.1-Terminus leads

Claude Sonnet 4.5: 32.3 (#216), DeepSeek-V3.1-Terminus: 38.5 (#137)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
LMArena Math14491402
FrontierMath (Tiers 1-3)23.9%—
FrontierMath Tier 42.4%—
OTIS Mock AIME 2024-202577.8%—
ProofBench19%—
Omni-MATH55.3%—
MATH Level 597.7%—
FrontierMath (Feb 2025 set)15.2%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge Not comparable

Claude Sonnet 4.5: 48.4 (#76), DeepSeek-V3.1-Terminus: —

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
GPQA Diamond82.3%—
Humanity's Last Exam13.7%—
SimpleQA Verified30.7%—
MMLU-Pro86.9%—
Vectara Hallucination Rate12%—
GPQA (HELM)68.6%—
LMArena Expert1482—

Multimodal Not comparable

Claude Sonnet 4.5: 34.8 (#89), DeepSeek-V3.1-Terminus: —

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
VPCT39.8%—
LMArena Document1450—

Multilingual Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 53.4 (#69), DeepSeek-V3.1-Terminus: 52.1 (#92)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
LMArena Non-English14251407
LMArena Russian14371436
LMArena Chinese1459—
LMArena French1458—
LMArena German1427—
LMArena Japanese1390—
LMArena Korean1403—
LMArena Spanish1457—

Instruction Following Too close to call

Claude Sonnet 4.5: 75.0 (#78), DeepSeek-V3.1-Terminus: 74.0 (#106)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
LMArena Instruction Following14591404
IFEval85%—

Long Context Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 45.2 (#46), DeepSeek-V3.1-Terminus: 43.4 (#97)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
LMArena Longer Query14761421

Writing & Preference Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 66.5 (#34), DeepSeek-V3.1-Terminus: 61.0 (#92)

BenchmarkClaude Sonnet 4.5DeepSeek-V3.1-Terminus
LMArena Text14391419
LMArena Creative Writing14421403
LMArena Multi-Turn14651411
EQ-Bench Creative Writing1678—
WildBench85.4%—

Frequently asked questions

Is Claude Sonnet 4.5 better than DeepSeek-V3.1-Terminus?

Claude Sonnet 4.5 and DeepSeek-V3.1-Terminus score almost the same on the Noometry Index (44.1 vs 43.1), so choose on price, context window or the category you care about most.

Which is cheaper, Claude Sonnet 4.5 or DeepSeek-V3.1-Terminus?

DeepSeek-V3.1-Terminus is cheaper. It lists at $0.27 per million input tokens and $1 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

Is Claude Sonnet 4.5 or DeepSeek-V3.1-Terminus better for coding?

Claude Sonnet 4.5 scores higher on coding benchmarks: 47.3 versus 42.0 in the Noometry coding category.

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-V3.1-Terminus share?

16 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and DeepSeek-V3.1-Terminus has 16.

Related comparisons

Go deeper