LLM Wiki is a cross-platform desktop application that turns your documents into an organized, interlinked knowledge base — automatically. Instead of traditional RAG (retrieve-and-answer from scratch every time), the LLM incrementally builds and maintains a persistent wiki from your sources。
TypeScript
Commits on main
fix: Gemini streaming — multi-part text + URL-encode model id
145e218 nash_su committed 7 months ago
chore: refresh Kimi/Zhipu/MiniMax model lists + merge dual-wire presets
45a1dc7 nash_su committed 7 months ago
chore: Volcengine Ark — replace model list with the coding-endpoint-only set
cc7fd9e nash_su committed 7 months ago
feat: add Alibaba Bailian Coding Plan presets (OpenAI + Anthropic wire)
184f7d2 nash_su committed 7 months ago
feat: route LLM fetch through Tauri HTTP plugin (bypass browser CORS)
cb1ebba nash_su committed 7 months ago
chore: Kimi presets — promote k2.6/k2.5, mark older models EOL 2026-05-25
1af4fda nash_su committed 7 months ago
fix: stop ghost "[Binary file: …]" outputs from overwriting real files
daacc42 nash_su committed 7 months ago
fix: web clips go through the ingest queue instead of direct autoIngest
e600a42 nash_su committed 7 months ago
build: commit PDFium binaries directly instead of fetching in CI
c120be9 nash_su committed 7 months ago
Merge branch 'pdf-pdfium': switch PDF extraction to pdfium-render
f57ba8c nash_su committed 7 months ago
fix: Windows path handling — normalize at FFI boundary and fix abs-path check
bcd4d2c nash_su committed 7 months ago
fix: search ranks exact matches first — filename/phrase bonuses dominate
da5c8df nash_su committed 7 months ago
test: verify panic_guard catches sync/async panics and tolerates malformed PDFs
5b028e2 nash_su committed 7 months ago
fix: catch panics in Tauri commands so one bad file can't kill the app
0a142d1 nash_su committed 7 months ago
fix: lower ingest LLM temperature to 0.1 for stable multi-file output
3536cae nash_su committed 7 months ago