DeepSeek-V3 · deepinfra

Name: DeepSeek-V3
Brand: DeepSeek
Price: 0.32 USD
Availability: InStock

DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2.

Modelo base Todos os modelos desta base Todos os modelos do provedor ao provedor

Entrada, $/1M $0.32

Saída, $/1M $0.89

Limites