Sebastian Raschka ↗ · 2025-12-03 20:03 · UTC+8
From DeepSeek V3 to V3.2: Architecture, Sparse Attention, and RL Updates
Media reports Sebastian Raschka Source language: English
Excerpt
Last updated: January 1st, 2026 Similar to DeepSeek V3, the team released their new flagship model over a major US holiday weekend. Given DeepSeek V3.2’s really good performance (on GPT-5 and Gemini 3.0 Pro) level, and the fact that it’s also available as an open-weight model, it’s definitely worth a closer look. Figure 1: Benchmark comparison between DeepSeek V3.2 and proprietary flagship models.
Short excerpt from the source — open the original for the full text.