Gemma 4 31B UD-Q8_K_XL Suddenly Hits "Exceeds Shared Memory" at 68k Context
Few_Professional6859 · reddit · 2026-09-26
A redditor reports that Gemma 4 31B (UD-Q8KXL quant) suddenly fails with an "Exceeds shared memory" error at 68k context on the exact same machine and setup that previously loaded and ran fine — suggesting a regression at the runtime/driver level rather than the model or hardware. Screenshot included. A notable bug for local long-context inference users.
More from Infra
- Running Qwen2.5-Coder-3B on Android: Jupyter + Pydantic Evals on-device eval pipeline — alex_b3b · 2026-11-11
- 22% of Nvidia revenue now books to Singapore, but only ~1% is local buyers — Gregory_C_Allen · 2026-11-11
- Huawei's Kirin 9050 NPU claims offline 30B MoE on smartphones, +140% perf — AIFlow_ML · 2026-11-11
- Stratechery: Intel vs AMD's diverging earnings and Intel's need for an old-new path — Stratechery · 2026-11-11
- a16z Podcast: Mike Volpi on What the 1990s Internet Boom Teaches Us About the AI Buildout — a16z Podcast · 2026-11-11
- Alibaba files to list chip-design unit T-Head in Hong Kong to fund AI push — pstAsiatech · 2026-11-11