DeepSeek V4 Flash 284B (13B active)

on 4× AMD RX 7900 XTX · llama.cpp · 1,048,576 ctx

Sep 7, 2026

Throughput

12.0 t/s gen · 141.4 tokens/s prompt processing (prefill)Prompt processing measures input prefill speed; it does not tell you the time to first token.

Quant

UD-IQ3_XXS

System RAM

128 GB

VRAM reported

120 GB

Summary

User reports prompt processing in the high 140s t/s, dropping to the mid 80s at 60k context, with generation at ~11.9 t/s. Setup uses the ROCm 7.2.4 backend on 1x 7900 XTX 24GB plus 3x MI60 32GB. One MI60 failed.

View source ↗