Key Specifications

Vendordeepseek
Versioncoder-33b
Release Date2024-01-25
Context Window16384 tokens
Input Modalitiestext
Output Modalitiestext
LicenseDeepSeek License
Documentationhttps://api-docs.deepseek.com/

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU65.9%2024-01-255-shotview
HUMANEVAL71.8pass@12024-01-25view
GSM8K69.7%2024-01-250-shot CoTview
MATH32.1%2024-01-250-shot CoTview
BBH61.6%2024-01-253-shot CoTview
GPQA26.8%2024-01-250-shotview
IFEVAL58.8%2024-01-25prompt_strictview
ARC92.2%2024-01-25challengeview
MUSR45.8%2024-01-250-shotview
WINOGRANDE79.8%2024-01-250-shotview

Pricing

TierPriceCurrency
Input$0.28 / MtokUSD
Output$0.28 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://api-docs.deepseek.com/quick_start/pricing · as of 2024-01-25

Compliance

  • Data Residency: CN
  • SOC2: ✗
  • HIPAA: ✗
  • GDPR: ✗
  • ISO 27001: ✗

DeepSeek Coder 33B

Visão geral do modelo

DeepSeek Coder 33B 代码专用开源模型, 16K 上下文, 33B 参数, 性能接近 GPT-3.5, 开源代码模型旗舰。

Especificações principais

FornecedorVersãoData de lançamentoJanela de contextoModalidades de entradaModalidades de saídaLicença
Deepseekcoder-33b2024-01-2516KtexttextDeepSeek License

Desempenho em benchmarks

BenchmarkPontuaçãoUnidadeNotas
MMLU (Massive Multitask Language Understanding)65.9%5-shot
HumanEval71.8pass@1
GSM8K (Grade School Math 8K)69.7%0-shot CoT
MATH32.1%0-shot CoT
BBH (BIG-Bench Hard)61.6%3-shot CoT
GPQA26.8%0-shot
IFEval58.8%prompt_strict
ARC92.2%challenge
MUSR45.8%0-shot
WinoGrande79.8%0-shot

Preços

EntradaSaídaLeitura de cacheEscrita de cache

por milhão de tokens

Pontos fortes

  • 可靠的通用模型。

Pontos fracos

  • 闭源专有模型,不支持自托管。
  • 上下文窗口 16K 偏小。

Casos de uso

  • 代码生成与调试

Referências