deepseek-ai/DeepSeek-VL2

DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding

36 best points · 7 comments · 3 Hacker News posts · 5,399 GitHub stars · View on GitHub

Hacker News discussions

• DeepSeek-VL2: MoE Vision-Language Models for Advanced Multimodal Understanding (36 points, 7 comments, by selvan, 2025-01-01)
• DeepSeek-VL2: Moe Vision-Language Models for Advanced Multimodal Understanding [pdf] (1 points, 0 comments, by limoce, 2024-12-13)
• DeepSeek VL2 (1 points, 0 comments, by arunbahl, 2024-12-16)