Selfslop Research & Systems Writing
⚠️ This is an engineering log, not a benchmark post. Every Phase A/B/C/D table row is "⬜ Not run" or "TBD." The async pipeline, MoE hot cache, and adaptive MTP changes described here have not been empirically validated end-to-end . The one measured number (71.46 t/s aggregate decode …
The release of Qwen3.6-27B in April 2026 sparked a flurry of experiments in the llama.cpp community. While full GPU offload remains the default for high-end inference, practitioners with partial offload hardware (single RTX 4090, consumer GPUs, laptops with iGPUs) have pushed for smarter tensor placement strategies. The paper …
The release of ChatGPT in November 2022 triggered the fastest technology-adoption curve in history. Within eight months, LLM-assisted content in professional writing had risen from a baseline of ~1.5% to over 15%. By mid-2023, a critical mass of practitioners had folded LLMs into their daily epistemic workflow: asking models …
There is a reproducible effect in large language models that most practitioners have observed but few have named: model-written prompts work better than human-written ones. This isn't about better prompt engineering in the human sense. It's about latent resonance — the alignment between a prompt's distributional signature and the target model's …