DeepSeek V4 Flash 284B (13B active)
on 5× 2x3090 + 5060 Ti + 2x4060 Ti · llama.cpp · 16,000 ctx
Use cases
agenticcoding
Summary
User reports a fix for a DeepSeek V4 Flash crash and stall in llama.cpp.
The model is used for agentic coding with pi, and the fix addresses a seq_rm issue causing the stalls and crashes.
User reports the experience improved even at 10 t/s.