AI search engine and MCP server for bots and agents. One Rust binary indexes code, docs, logs, JSONL/CSV and PDFs for BM25 full-text, vector and hybrid search, code search without grep, agent memory and RAG. Elasticsearch-compatible API; Corpus Hub serves hundreds of reference memory corpora.
Rust agent-memoryai-agentsbm25claude-codeclicode-searchcursordatabaseelasticsearchfull-text-searchhybrid-searchjsonllocal-firstmcpmcp-servermodel-context-protocolragrustsearch-enginevector-search
docs: correct the binary-size claim with measured figures
xerj-org0d7e201dfd6d
diff --git a/docs/WHY_XERJ.md b/docs/WHY_XERJ.mdindex d13320cd7..45e1d629a 100644--- a/docs/WHY_XERJ.md+++ b/docs/WHY_XERJ.md@@ -25,7 +25,7 @@ The autoindex pipeline was verified adversarially: 80/81 ground-truth checks on## Engineering positions (where XERJ deliberately differs)-- **One static binary (~36 MB, neural embedder included), no JVM.** Sub-second cold start. An agent can spawn its own search engine as a subprocess tool. Operationally: no heap tuning, no GC pauses. (A `--no-default-features` slim build without the neural backend is ~23 MB.)+- **One static binary (61.6 MB, neural embedder included), no JVM.** Sub-second cold start. An agent can spawn its own search engine as a subprocess tool. Operationally: no heap tuning, no GC pauses. (A `--no-default-features` slim build without the neural backend is 58.2 MB — the neural backend accounts for 3.4 MB of the total, so dropping it is rarely worth the loss of capability.)- **Columnar aggregation fast paths + doc-value prefilters.** The analytical questions agents actually ask (counts, cardinalities, histograms, percentiles over millions of rows) are served from columnar data, not stored-document scans. This is where XERJ *beats* Elasticsearch in the head-to-head.- **HNSW-served kNN with exact rescoring, measured honestly.** Unfiltered kNN is served by a persisted HNSW graph whose candidates are exact-rescored, so returned scores match the exact path bit-for-bit — measured recall@10 1.00 on the official bench query, 100-probe mean 0.976 (ES 8.13.4 same protocol: 0.937). Filtered, non-cosine, SQ8-quantized, and small-index shapes keep the exact brute-force scan (recall 1.00 by construction), and the trade-offs are documented rather than hidden behind approximate defaults.- **Built-in zero-config embedder.** Lexical feature-hashing (384-dim) — documented honestly as *hybrid lexical+vector*, never claimed to be neural. Semantic-ish retrieval works offline with zero external dependencies; point `semantic_text` at an external inference endpoint when you need real embeddings.