vLLM @vllm_project · 01 Sep 2026
🎉 Congrats to @deepseek_ai on DeepSeek-V4-Flash-Vision-Exp, the first multimodal model in the V4 family! vLLM serves it now. A vision encoder and aligner on the 285B/13B V4-Flash MoE backbone, with text-agent scores holding steady. 🚀 🔗 https://t.co/sxl7k5w9rs https://t.co/X162SYzoW2
18 768Views
253Likes
27Reposts
16Replies
1Quotes
47Bookmarks
Is that a lot?
0.72×vs this author's median25 968 views is typical
33Percentile for this authorof 6 recent posts
6.82×vs 10K–100K median2 751 views is typical
38.81%Reachviews ÷ followers
1.58%Engagement rateof viewers reacted