D

DeepSeek: deepseek-v3-0324

deepseek-v3-0324
128K context8.2K outReasoningToolsFiles
Released Mar 24, 2025Knowledge cutoff Aug 2025Updated Aug 20, 2026

A state-of-the-art 671B MoE model (37B active) featuring Multi-head Latent Attention (MLA) and auxiliary-loss-free load balancing. It provides a hybrid "Think/Non-think" mode, delivering SOTA performance in coding, math, and agentic tasks while remaining the most cost-efficient flagship in the 2026 market.

Mode chatDeprecation Jul 2026

Pricing

Input price
$0.50/ 1M tokens
Output price
$2.00/ 1M tokens
Context window 163.8K tokensCompatible endpoints openaiVendor DeepSeek

Uptime

Performance

Loading performance data...

Usage & Ranking

Loading usage...

Frequently asked questions

Similar models