Noometry

Model comparison

Claude 3.5 Haiku vs DeepSeek-V3

DeepSeek-V3 is the stronger model overall, scoring 39.5 to 29.2 on the Noometry Index.

Last verified . 45 shared benchmarks.

Claude 3.5 Haiku Anthropic

29.2

Rank #315 Confirmed

DeepSeek-V3 DeepSeek

39.5

Rank #166 Confirmed

Summary

  • They share 45 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 1 category and DeepSeek-V3 in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where DeepSeek-V3 leads 37.5 to 18.7.
  • The biggest single-benchmark swing is LiveBench Math: 35.5% for Claude 3.5 Haiku and 73.5% for DeepSeek-V3.
  • DeepSeek-V3 has downloadable open weights; the other is API-only.

Side by side

Claude 3.5 HaikuDeepSeek-V3
ProviderAnthropicDeepSeek
Noometry Index29.239.5
Released2024-10-222024-12-26
WeightsProprietaryOpen
Context window—164K
Max output—164K
Input $ / M tokens—$0.24
Output $ / M tokens—$0.90
Results tracked4960

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

Category by category

Coding DeepSeek-V3 leads

Claude 3.5 Haiku: 32.9 (#265), DeepSeek-V3: 42.3 (#106)

BenchmarkClaude 3.5 HaikuDeepSeek-V3
Aider Polyglot28%55.1%
SciCode27.4%35.8%
WeirdML30.7%36.1%
BigCodeBench Instruct46.1%50%
LiveBench Coding51.4%70.9%
LMArena Coding12861368
BigCodeBench Complete59%62.2%
CadEval32%—
HumanEval+—86.6%
MBPP+—73%

Agentic & Tool Use Not comparable

Claude 3.5 Haiku: 28.0 (#95), DeepSeek-V3: —

BenchmarkClaude 3.5 HaikuDeepSeek-V3
BALROG19.3%—
METR Time Horizons—49.6%

Reasoning DeepSeek-V3 leads

Claude 3.5 Haiku: 17.7 (#290), DeepSeek-V3: 20.5 (#236)

BenchmarkClaude 3.5 HaikuDeepSeek-V3
CritPt0%0%
LiveBench Reasoning28.1%65.8%
LMArena Hard Prompts12511365
DTBench56.7%64.8%
LiveBench Data Analysis48.5%60.9%
Epoch Capabilities Index127.15135.94
LiveBench43.5%66.9%
SimpleBench—27.2%
Kagi LLM Benchmark—52.3%
LMCA—15.5%
BIG-Bench Hard—87.5%
ForecastBench—59.1
HellaSwag—88.9%
PIQA—84.7%
WinoGrande—85.2%

Math DeepSeek-V3 leads

Claude 3.5 Haiku: 14.7 (#300), DeepSeek-V3: 32.1 (#219)

BenchmarkClaude 3.5 HaikuDeepSeek-V3
OTIS Mock AIME 2024-20254.3%37.8%
Omni-MATH22.4%40.3%
LiveBench Math35.5%73.5%
LMArena Math12441373
MATH Level 546.4%75.5%
FrontierMath (Feb 2025 set)0.3%1.7%

Knowledge DeepSeek-V3 leads

Claude 3.5 Haiku: 18.7 (#281), DeepSeek-V3: 37.5 (#155)

BenchmarkClaude 3.5 HaikuDeepSeek-V3
GPQA Diamond38.1%67.6%
MMLU-Pro60.5%72.3%
Confabulations36.7%26.1%
GPQA (HELM)36.3%53.8%
LMArena Expert12081351
MMLU74.3%87.2%
Vectara Hallucination Rate—6.1%
ARC (AI2) Challenge—95.3%
TriviaQA—82.9%

Multimodal Not comparable

Claude 3.5 Haiku: 26.8 (#117), DeepSeek-V3: —

BenchmarkClaude 3.5 HaikuDeepSeek-V3
LMArena Vision1092—
GeoBench34%—

Multilingual DeepSeek-V3 leads

Claude 3.5 Haiku: 40.0 (#218), DeepSeek-V3: 48.5 (#143)

BenchmarkClaude 3.5 HaikuDeepSeek-V3
LMArena Non-English12381358
LMArena Chinese12291391
LMArena French12641385
LMArena German12371374
LMArena Japanese11751333
LMArena Korean11731319
LMArena Russian12531373
LMArena Spanish12611358

Instruction Following DeepSeek-V3 leads

Claude 3.5 Haiku: 62.9 (#234), DeepSeek-V3: 72.8 (#130)

BenchmarkClaude 3.5 HaikuDeepSeek-V3
LiveBench Instruction Following61.9%81.5%
IFEval79.2%83.2%
LMArena Instruction Following12411345

Long Context Claude 3.5 Haiku leads

Claude 3.5 Haiku: 38.3 (#200), DeepSeek-V3: 34.0 (#253)

BenchmarkClaude 3.5 HaikuDeepSeek-V3
LMArena Longer Query12611352
Fiction.LiveBench—50%

Writing & Preference DeepSeek-V3 leads

Claude 3.5 Haiku: 42.7 (#234), DeepSeek-V3: 57.4 (#130)

BenchmarkClaude 3.5 HaikuDeepSeek-V3
LMArena Text12551375
LMArena Creative Writing12331364
Short-Story Creative Writing73.5%77%
EQ-Bench Creative Writing11461472
WildBench76%83%
LMArena Multi-Turn12651389
LiveBench Language35.4%49.1%

Frequently asked questions

Is Claude 3.5 Haiku better than DeepSeek-V3?

DeepSeek-V3 is the stronger model overall, scoring 39.5 to 29.2 on the Noometry Index.

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

DeepSeek-V3 scores higher on coding benchmarks: 42.3 versus 32.9 in the Noometry coding category.

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

45 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and DeepSeek-V3 has 60.

Related comparisons

Go deeper