Noometry

Model comparison

Claude 3.5 Haiku vs DeepSeek-V3.1

DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 28.9 on the Noometry Index.

Last verified . 21 shared benchmarks.

Claude 3.5 Haiku Anthropic

28.9

Rank #309 Confirmed

DeepSeek-V3.1 DeepSeek

42.2

Rank #106 Confirmed

Summary

  • They share 21 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 1 category and DeepSeek-V3.1 in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.0 to 18.7.
  • The biggest single-benchmark swing is DTBench: 56.7% for Claude 3.5 Haiku and 82.7% for DeepSeek-V3.1.
  • DeepSeek-V3.1 has downloadable open weights; the other is API-only.

Side by side

Claude 3.5 HaikuDeepSeek-V3.1
ProviderAnthropicDeepSeek
Noometry Index28.942.2
Released2024-10-222025-08-21
WeightsProprietaryOpen
Context window—164K
Max output—8K
Input $ / M tokens—$0.25
Output $ / M tokens—$0.95
Results tracked4927

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

Category by category

Coding DeepSeek-V3.1 leads

Claude 3.5 Haiku: 32.6 (#258), DeepSeek-V3.1: 39.8 (#140)

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
WeirdML30.7%38.4%
LMArena Coding12871417
Aider Polyglot28%—
SciCode27.4%—
BigCodeBench Instruct46.1%—
LiveBench Coding51.4%—
BigCodeBench Complete59%—
CadEval32%—

Agentic & Tool Use Not comparable

Claude 3.5 Haiku: 27.7 (#90), DeepSeek-V3.1: —

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
BALROG19.3%—

Reasoning DeepSeek-V3.1 leads

Claude 3.5 Haiku: 17.1 (#286), DeepSeek-V3.1: 27.3 (#109)

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
LMArena Hard Prompts12511417
DTBench56.7%82.7%
Epoch Capabilities Index127.15139.92
SimpleBench—40%
Kagi LLM Benchmark—53.2%
CritPt0%—
LiveBench Reasoning28.1%—
LiveBench Data Analysis48.5%—
LMCA—24.3%
ForecastBench—58
LiveBench43.5%—

Math DeepSeek-V3.1 leads

Claude 3.5 Haiku: 13.7 (#294), DeepSeek-V3.1: 36.1 (#125)

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
LMArena Math12441420
OTIS Mock AIME 2024-20254.3%—
Omni-MATH22.4%—
LiveBench Math35.5%—
MATH Level 546.4%—
FrontierMath (Feb 2025 set)0.3%—

Knowledge DeepSeek-V3.1 leads

Claude 3.5 Haiku: 18.7 (#276), DeepSeek-V3.1: 43.0 (#86)

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
LMArena Expert12081403
GPQA Diamond38.1%—
MMLU-Pro60.5%—
Confabulations36.7%—
Vectara Hallucination Rate—5.5%
GPQA (HELM)36.3%—
MMLU74.3%—

Multimodal Not comparable

Claude 3.5 Haiku: 26.3 (#113), DeepSeek-V3.1: —

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
LMArena Vision1092—
GeoBench34%—

Multilingual DeepSeek-V3.1 leads

Claude 3.5 Haiku: 40.2 (#213), DeepSeek-V3.1: 51.8 (#100)

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
LMArena Non-English12381400
LMArena Chinese12291470
LMArena French12621447
LMArena German12371412
LMArena Japanese11751374
LMArena Korean11721337
LMArena Russian12531405
LMArena Spanish12611429

Instruction Following DeepSeek-V3.1 leads

Claude 3.5 Haiku: 63.1 (#229), DeepSeek-V3.1: 74.1 (#105)

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
LMArena Instruction Following12411400
LiveBench Instruction Following61.9%—
IFEval79.2%—

Long Context Claude 3.5 Haiku leads

Claude 3.5 Haiku: 38.3 (#195), DeepSeek-V3.1: 36.3 (#227)

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
LMArena Longer Query12611422
Fiction.LiveBench—52.8%

Writing & Preference DeepSeek-V3.1 leads

Claude 3.5 Haiku: 42.9 (#228), DeepSeek-V3.1: 60.5 (#93)

BenchmarkClaude 3.5 HaikuDeepSeek-V3.1
LMArena Text12551420
LMArena Creative Writing12331401
EQ-Bench Creative Writing11461436
LMArena Multi-Turn12651408
Short-Story Creative Writing73.5%—
WildBench76%—
LiveBench Language35.4%—

Frequently asked questions

Is Claude 3.5 Haiku better than DeepSeek-V3.1?

DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 28.9 on the Noometry Index.

Is Claude 3.5 Haiku or DeepSeek-V3.1 better for coding?

DeepSeek-V3.1 scores higher on coding benchmarks: 39.8 versus 32.6 in the Noometry coding category.

How many benchmarks do Claude 3.5 Haiku and DeepSeek-V3.1 share?

21 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and DeepSeek-V3.1 has 27.

Related comparisons

Go deeper