qmd · Issues· 149 open

vsearch takes ~70 s on an 11k-vector collection: exact scan runs per expanded query, expansions not deduplicated

doctor cache check misses revision-pinned HF models that resolve offline

BUG: qmd search and qmd get serve deleted files (and stale content) silently

exclude: patterns in index.yml to prevent double-indexing nested collections

Would a minimal, read-only GUI front-end for qmd collections be useful to anyone?

MCP get/multi_get return embedded-resource content; text-only clients render garbage

`qmd embed` still fails mid-run with DisposedError on 2.8.3, and still exits 0

createEmbeddingContext failure doesn't fall back to CPU the way GPU backend init does

Bug report: chunker splits markdown tables mid-rows

[BUG, WINDOWS] mcp daemon is not stopped on qmd mcp stop

Embedding inputs at/over the context's physical batch boundary (512 tokens) silently store deterministic garbage vectors

HTTP MCP daemon disposes active LLM contexts after five minutes, causing DisposedError and SIGSEGV

Feature Request: Allow remote model selection for Docker Model Runner support

bench: pathsMatch is an unanchored suffix test — a fixture naming a file that does not exist scores f1 1.0

Concurrent qmd invocations each load models independently — exhausts system RAM on CPU-only hosts (refiling #147 with 2.8.3 repro)