deepseek-v4.1POST

DeepSeek V4.1-Flash-1017 runs on 2x DGX Spark at EXL3 2.9bpw

A ~210GB EXL3 quant with only about 2% code perplexity loss drops into the existing V4.1 Spark kit, though the new 1017 Engram tables must be pulled.

𝙒𝙞𝙡𝙘𝙤 𝘷𝘢𝘯 𝘽𝙤𝙭𝙩𝙚𝙡 ⚡🚀🧲🌞

@nuonrgArchived 11 days agoOriginally posted 2026.10.17

𝙒𝙞𝙡𝙘𝙤 𝘷𝘢𝘯 𝘽𝙤𝙭𝙩𝙚𝙡 ⚡🚀🧲🌞 @nuonrgOriginally posted 2026.10.17

Open original on X

DeepSeek V4.1 Flash 1017 is now runnable on 2x DGX Spark in EXL3: a 2.9 bpw quant (~210 GB) with only ~2% perplexity loss on code, drop-in for the existing V4.1 Spark kit. Remember to pull the new 1017 Engram tables, the old ones changed. #AI #Deepseek #LocalAI https://huggingface.co/LibertAIDAI/DeepSeek-V4.1-Flash-1017-EXL3-2.9bpw

x.com/nuonrg/status/2111473990065590666Archived 2026.10.18 16:45