Tag

# KV cache compression

1 post with this tag.

Oct 18, 20263 min readAI ModelsAI generated

DeepSeek V4.1 Flash-1017 Released

DeepSeek V4.1 Flash-1017 targets reasoning, coding, agents, and vision with shorter execution, but its deployment cost remains substantial.

DeepSeek V4.1 Flash-1017 DeepSeek multimodal model KV cache compression Hugging Face