DeepSeek: DeepSeek V3.2

deepseek/deepseek-v3.2
Created Dec 1, 2025|128K context context|Starting at $3/M input tokens|Starting at $3/M output tokens

DeepSeek V3.2 is a reasoning-oriented DeepSeek model designed for efficient long-context processing, coding, and agentic tool-use workloads. It uses sparse-attention style optimizations to improve throughput while preserving strong performance on multi-step reasoning and software tasks.

Providers for DeepSeek V3.2

Routes requests to the best providers that are able to handle your prompt size and parameters.

ProviderContext WindowMax OutputInput PriceOutput PriceCache ReadCache Write
Volcengine
128K32K
$3
$3
$0.0252