| .. |
|
bench
|
b7a5a86a9b
feat(cli): add `qmd bench` command for search quality benchmarks
|
4 meses atrás |
|
cli
|
c8d49d26da
feat(cli): configure LlamaCpp singleton from YAML models config
|
4 meses atrás |
|
mcp
|
ce0cd64409
feat(mcp): pass YAML config path to createStore for model resolution
|
4 meses atrás |
|
ast.ts
|
244ddf5ecb
feat: AST-aware chunking for code files via tree-sitter
|
5 meses atrás |
|
bench-rerank.ts
|
3095041e0f
feat: use `build: "autoAttempt"` on `getLlama`
|
5 meses atrás |
|
collections.ts
|
14b384d9d8
feat(config): add ModelsConfig type to CollectionConfig
|
4 meses atrás |
|
db.ts
|
52ebafce11
fix(db): add macOS Homebrew SQLite support for Bun and restore actionable errors
|
5 meses atrás |
|
embedded-skills.ts
|
b16d77146a
feat(skill): install packaged qmd skill
|
5 meses atrás |
|
index.ts
|
33d42a2a04
feat(sdk): pass YAML models config to LlamaCpp in createStore
|
4 meses atrás |
|
llm.ts
|
8644fa99d1
fix(store): thread embed model URI to format functions for correct prompt detection
|
4 meses atrás |
|
maintenance.ts
|
c68904fe08
refactor: move CLI and MCP to subdirectories, MCP consumes SDK
|
5 meses atrás |
|
store.ts
|
8644fa99d1
fix(store): thread embed model URI to format functions for correct prompt detection
|
4 meses atrás |
|
test-preload.ts
|
537d15a9e6
fix: proper cleanup of Metal GPU resources in tests
|
6 meses atrás |
JohnRichardEnders
8644fa99d1
fix(store): thread embed model URI to format functions for correct prompt detection