Qwen3.8 125B (6B active) Swift-1.5

on Strix Halo 64GB · Strata

Oct 8, 2026

Throughput

45.0 t/s gen · 500.0 tokens/s prompt processing (prefill)Prompt processing measures input prefill speed; it does not tell you the time to first token.

Quant

IQ2_XS (GGUF)

System RAM

64 GB

Summary

User reports running Swift-1.5-Qwen3.8-Flash-Next at about 500 t/s prompt processing and 45 t/s generation on a 64GB Strix Halo box. Setup is Strata with an IQ2_XS GGUF quant on Linux (NixOS), with a lot of memory left free. IQ3_S was tried but decoding was 10 times slower. The user calls it an upgrade over gufo and llama.cpp, and has not tried halogen.

View source ↗