RaymondHuang210129/llama.cpp-adaptive-kv-streaming

LLM inference in C/C++

5 best points · 2 comments · 1 Hacker News post · 472 GitHub stars · View on GitHub

Hacker News discussions

• The Llama.cpp Fork That Enables Qwen 3.8 27B Large Contexts for 16GB VRAM GPU (5 points, 2 comments, by dazhbog, 2026-08-31)