DeepSeek V4 Flash 284B (13B active)
on Unknown GPU · llama.cpp
Summary
User reports DeepSeek V4 Flash quantized to a custom 3-bit with llama.cpp PR #24162 running at 5-6 t/s.
The run is at an early stage, with stability and performance tradeoffs still being worked out.
User praises the model's intelligence, its efficiency as context length scales, and its low KV cache usage, and predicts it will dominate the 80-140 GB model space.