Model comparison

Claude Instant vs DeepSeek Coder 6.7B

Neither Claude Instant nor DeepSeek Coder 6.7B has enough public benchmark results to be ranked yet; the rows below show what has been published.

Last verified . 5 shared benchmarks.

Claude Instant Anthropic

29.5

Unranked Sparse

DeepSeek Coder 6.7B DeepSeek

37.6

Unranked Sparse

Summary

  • They share 5 benchmarks with published results for both.
  • DeepSeek Coder 6.7B has downloadable open weights; the other is API-only.

Side by side

Claude Instant and DeepSeek Coder 6.7B specifications
Claude InstantDeepSeek Coder 6.7B
ProviderAnthropicDeepSeek
Noometry Index29.537.6
Released2023-08-092023-11-02
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked79

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

Category by category

Coding Not comparable

Claude Instant: —, DeepSeek Coder 6.7B: 35.8 (#219)

Coding benchmarks
BenchmarkClaude InstantDeepSeek Coder 6.7B
HumanEval+50.6%71.3%
BigCodeBench Instruct—35.5%
BigCodeBench Complete—43.8%
MBPP+—65.6%

Reasoning Not comparable

Claude Instant: 19.6, DeepSeek Coder 6.7B: —

Reasoning benchmarks
BenchmarkClaude InstantDeepSeek Coder 6.7B
Epoch Capabilities Index120.2889.62
DTBench45.8%—
WinoGrande—57.6%

Math Not comparable

Claude Instant: —, DeepSeek Coder 6.7B: —

Math benchmarks
BenchmarkClaude InstantDeepSeek Coder 6.7B
GSM8K86.7%21.3%

Knowledge Not comparable

Claude Instant: —, DeepSeek Coder 6.7B: —

Knowledge benchmarks
BenchmarkClaude InstantDeepSeek Coder 6.7B
ARC (AI2) Challenge86.3%36.4%
MMLU73.4%36.4%
TriviaQA78.9%—

Frequently asked questions

Is Claude Instant better than DeepSeek Coder 6.7B?

Neither Claude Instant nor DeepSeek Coder 6.7B has enough public benchmark results to be ranked yet; the rows below show what has been published.

How many benchmarks do Claude Instant and DeepSeek Coder 6.7B share?

5 benchmarks have published results for both models. Claude Instant has 7 scored results on Noometry and DeepSeek Coder 6.7B has 9.

Related comparisons

Go deeper