Local LLMs
wire
2026-10-19
sglang v0.5.22 lands, MLX fixes M1/M2 SDPA
- v0.5.22 sgl-project/sglang
$ git log --author="ch-wan" --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.
13
stories filed
8
daily wires
$ grep -rn "author: ch-wan" wires/ | sort -r
Local LLMs
wire
2026-08-30
llama.cpp Vulkan memory bloat fixed, vLLM deprecates PyAV
- config: the resolution pipeline's dispatcher leaves the record sgl-project/sglang
Local LLMs
wire
2026-08-16
LLAMA.CPP OVERHAULS INFERENCE FLAGS, VLLM PATCHES DEEPSEEK SPARSE MLA
Local LLMs
wire
2026-08-11
OLLAMA PATCHES CRITICAL PARSER BUG, NEMOTRON 3 LANDS WITH SPECULATIVE DECODING
Local LLMs
wire
2026-08-10
OLLAMA SHIPS VISION SUPPORT ACROSS MLX AND QWEN3.5 AS LLAMA.CPP FIXES DISPATCH BUGS
- config: the KV-cache configurator reads the bags sgl-project/sglang
- config: the runner and scheduler read resolved config from the bags sgl-project/sglang
Local LLMs
wire
2026-08-02
LLAMA.CPP SHIPS FOUR BUILDS IN 24 HOURS AS VLLM AND SGLANG RACE ON SEQUENCE PARALLELISM
- config: route parallel config-leaf reads through get_parallel() (#33170) sgl-project/sglang
- test: recover the config-namespace-migration deferrals (#33171) sgl-project/sglang
Local LLMs
wire
2026-08-01
QWEN3.5 SPECULATIVE DECODING HITS 70% SPEEDUP ON LLAMA.CPP, OLLAMA ROUTES CLOUD MODELS
- [unified-memory] Let Kimi-Linear use the paged MLA attention backends sgl-project/sglang
Local LLMs
wire
2026-07-23
OLLAMA AGENTS NOW IMPORT SKILLS FROM CODEX AND CLAUDE
- runtime_context: add resolved-config namespace bags and accessors sgl-project/sglang
- Revert RuntimeContext config-namespace reads/roles (#31813–#31817) sgl-project/sglang