Local LLMs
wire
2026-10-31
Deterministic top-p under TP, KV constants stay on the GPU
- cmake : embed an app manifest in all Windows exes ggml-org/llama.cpp
$ git log --author="angt" --all --oneline
Every morning, RepoJournal's newsroom reads what shipped across the open-source projects developers depend on — and writes the wire. Your commits kept making the news. This is your clippings file: the stories where your work was the story.
5
stories filed
5
daily wires
$ grep -rn "author: angt" wires/ | sort -r
Local LLMs
wire
2026-09-13
vLLM EngineCore can be killed by crafted stop_token_ids, llama.cpp reworks JSON schema handling
- ui : add cache ggml-org/llama.cpp
Local LLMs
wire
2026-09-06
vLLM closes scale-out multimodal handoff hole
- ui : embed assets directly with CMake ggml-org/llama.cpp
Local LLMs
wire
2026-09-05
vLLM Fast Start maps weights zero-copy; GLMGA sampling capped
- ggml : don't crash when backend search path can't be read (#28271) ggml-org/llama.cpp
Local LLMs
wire
2026-07-26
LLAMA.CPP LOCKS DOWN WASM, ADDS MCP STDIO, HEXAGON GETS IM2COL ACCELERATION
- vendor : update cpp-httplib to 0.51.0 ggml-org/llama.cpp