Noometry

Model comparison

Claude 3 Sonnet vs Deepseek Coder v2

Deepseek Coder v2 is the stronger model overall, scoring 35.3 to 28.8 on the Noometry Index.

Last verified . 22 shared benchmarks.

Claude 3 Sonnet Anthropic

28.8

Rank #311 Confirmed

Deepseek Coder v2 DeepSeek

35.3

Rank #217 Confirmed

Summary

  • They share 22 benchmarks with published results for both. Claude 3 Sonnet scores higher in 3 categories and Deepseek Coder v2 in 5 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where Deepseek Coder v2 leads 32.3 to 10.0.
  • The biggest single-benchmark swing is BigCodeBench Complete: 53.8% for Claude 3 Sonnet and 59.7% for Deepseek Coder v2.
  • Deepseek Coder v2 has downloadable open weights; the other is API-only.

Side by side

Claude 3 SonnetDeepseek Coder v2
ProviderAnthropicDeepSeek
Noometry Index28.835.3
Released2024-02-292024-06-17
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked3024

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

Category by category

Coding Deepseek Coder v2 leads

Claude 3 Sonnet: 29.2 (#296), Deepseek Coder v2: 37.5 (#178)

BenchmarkClaude 3 SonnetDeepseek Coder v2
BigCodeBench Instruct42.7%48.2%
LMArena Coding12231251
BigCodeBench Complete53.8%59.7%
HumanEval+64%82.3%
MBPP+69.3%75.1%
WeirdML10.2%—

Reasoning Deepseek Coder v2 leads

Claude 3 Sonnet: 19.7 (#232), Deepseek Coder v2: 22.6 (#175)

BenchmarkClaude 3 SonnetDeepseek Coder v2
LMArena Hard Prompts11971207
WinoGrande75.1%83.7%
DTBench53.6%—
Epoch Capabilities Index120.69—

Math Deepseek Coder v2 leads

Claude 3 Sonnet: 10.0 (#305), Deepseek Coder v2: 32.3 (#188)

BenchmarkClaude 3 SonnetDeepseek Coder v2
LMArena Math12131241
OTIS Mock AIME 2024-20252.5%—
MATH Level 518.2%—
GSM8K—94.5%

Knowledge Deepseek Coder v2 leads

Claude 3 Sonnet: 21.1 (#271), Deepseek Coder v2: 31.7 (#209)

BenchmarkClaude 3 SonnetDeepseek Coder v2
LMArena Expert11731181
GPQA Diamond40.6%—
ARC (AI2) Challenge—64.3%
MMLU75.9%—

Multimodal Not comparable

Claude 3 Sonnet: 24.7 (#120), Deepseek Coder v2: —

BenchmarkClaude 3 SonnetDeepseek Coder v2
LMArena Vision984—

Multilingual Claude 3 Sonnet leads

Claude 3 Sonnet: 38.0 (#229), Deepseek Coder v2: 36.4 (#235)

BenchmarkClaude 3 SonnetDeepseek Coder v2
LMArena Non-English12051182
LMArena Chinese11881201
LMArena French12281184
LMArena German12041164
LMArena Japanese11311126
LMArena Korean11281104
LMArena Russian12271188
LMArena Spanish12041153

Instruction Following Claude 3 Sonnet leads

Claude 3 Sonnet: 63.1 (#230), Deepseek Coder v2: 61.9 (#237)

BenchmarkClaude 3 SonnetDeepseek Coder v2
LMArena Instruction Following12001180

Long Context Too close to call

Claude 3 Sonnet: 36.7 (#223), Deepseek Coder v2: 37.0 (#219)

BenchmarkClaude 3 SonnetDeepseek Coder v2
LMArena Longer Query12111219

Writing & Preference Claude 3 Sonnet leads

Claude 3 Sonnet: 42.3 (#233), Deepseek Coder v2: 38.4 (#249)

BenchmarkClaude 3 SonnetDeepseek Coder v2
LMArena Text12181191
LMArena Creative Writing11861120
LMArena Multi-Turn12271176

Frequently asked questions

Is Claude 3 Sonnet better than Deepseek Coder v2?

Deepseek Coder v2 is the stronger model overall, scoring 35.3 to 28.8 on the Noometry Index.

Is Claude 3 Sonnet or Deepseek Coder v2 better for coding?

Deepseek Coder v2 scores higher on coding benchmarks: 37.5 versus 29.2 in the Noometry coding category.

How many benchmarks do Claude 3 Sonnet and Deepseek Coder v2 share?

22 benchmarks have published results for both models. Claude 3 Sonnet has 30 scored results on Noometry and Deepseek Coder v2 has 24.

Related comparisons

Go deeper