deepseek-v4.1POST
Local deployment notes: V4.1-Flash-1017 weights are 510GB, 204GB of it Engram
A user inspecting the Hugging Face weights estimates 510GB total with 204GB in Engram, targeting around 30 tok/s if trimmed to 306GB.
となりのトトノ🏯Local LLM | Tonoken3
@Tono_Ken3Archived 10 days agoOriginally posted 2026.10.17
となりのトトノ🏯Local LLM | Tonoken3 @Tono_Ken3Originally posted 2026.10.17
あれ?これってもう来てるんじゃんね😆 重み510GBうち204GBがEngram 306GBなら目標は大きく30TPSといったところだな。よい https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017
x.com/tono_ken3/status/2111374445529215344Archived 2026.10.18 16:45