Hacker News · item history

Qwen3.8-Flash-Next

207

first seen points

671

peak points

671

latest points

observations

first seen Aug 26, 2026 · last seen Aug 27, 2026 · published Aug 26, 2026

Posted by	tosh

Featured on 1 day: August 27, 2026.

Posted on X

2026-08-27
Qwen3.8-Flash-Next benchmarks at ~2x the speed of GPT-4o mini on standard evals. Speed matters in production — until the bill shows up. The pricing page doesn't list per-task cost yet, which is the only number that tells you if 2x faster means half the spend.