Qwen3.8-Flash-Next may be the best open model on 128GB unified memory
victormustar · x · 2026-09-03
Victor Mustar suggests Qwen3.8-Flash-Next is probably the strongest open model currently runnable on 128GB of unified memory, calling it "extremely strong" and planning to test it this week.
It's an early impression rather than a full benchmark, but signals a new bar for open-weight models on large-memory local hardware.
More from Models
- Meta Muse Spark 1.3 early coding tests impress: near-frontier quality at a third of the price — DeryaTR_ · 2026-09-03
- Qwen3.8 Flash AP quants beat other high-quality quants with new KLD eval method — Dutchnamn · 2026-09-03
- GLM-OCR reads Nvidia's 61-page 10-Q at 2,086 tok/s for under 2 cents — spillai · 2026-09-03
- Alexandr Wang Touts Meta Muse Spark 1.3: 1 Minute vs Fable 5.1's 70 Minutes and $13 — alexandr_wang · 2026-09-03
- A Gemini Flash model reportedly tops the DeepSWE coding leaderboard — sunjiao123sun_ · 2026-09-03
- Hypothesis: the better LLMs get at coding, the worse their writing gets — kwangmoo_yi · 2026-09-03