Commit Graph

71 Commits

Author SHA1 Message Date
shenlan
8f23398219 refactor askai to support local and online modes 2025-08-14 08:52:54 +08:00
shenlan
f341beaf9f Merge pull request #166 from svc-design/codex/verify-rag-implementation-in-askai-component
docs: document AskAIDialog RAG flow
2025-08-13 17:42:36 +08:00
shenlan
efff407be5 docs: document AskAIDialog RAG flow 2025-08-13 17:23:11 +08:00
Haitao Pan
0511501386 docs: fix zhparser build instructions for macOS Homebrew setup 2025-08-13 16:27:33 +08:00
Haitao Pan
d306ad8f00 docs: enhance pg-vector-db-init.md with zhparser installation steps 2025-08-13 15:15:41 +08:00
Haitao Pan
e5fe62d009 update: docs/pg-vector-db-init.md 2025-08-13 14:53:32 +08:00
shenlan
a479b10efd feat(rag): add zhcn schema 2025-08-13 14:00:57 +08:00
shenlan
9c50d22189 Merge pull request #162 from svc-design/codex/simplify-init.sql-for-hybrid-search
Simplify RAG DB schema for hybrid search
2025-08-13 13:17:23 +08:00
shenlan
ec94825fbe Simplify RAG schema and upsert logic 2025-08-13 13:14:35 +08:00
Haitao Pan
b4d7ccb241 update docs/models.txt 2025-08-13 13:12:34 +08:00
Haitao Pan
84cb83933d Add benchmarking scripts and configs:
- bench_embedding.sh / bench_ollama.sh for Ollama & embedding API tests
- hf_embedding_bench.py for HF model performance
- models.txt / models-emb.txt for test configs
- docs in bench_embedding.md / bench_ollama.md
2025-08-13 13:12:34 +08:00
shenlan
8680731e1d Merge pull request #158 from svc-design/codex/add-best-practices-for-document-qa-embedding
docs: add QA embedding best practices
2025-08-13 10:40:41 +08:00
shenlan
eaf0d01d04 docs: add QA embedding best practices 2025-08-13 10:39:15 +08:00
shenlan
5676a11927 feat(rag): enhance chunking and embedding 2025-08-13 09:50:13 +08:00
shenlan
61d4dff7d3 Merge branch 'main' into codex/optimize-cli-server-with-langchaingo 2025-08-12 23:48:45 +08:00
shenlan
96eba3c71e fix: correct generator endpoint handling 2025-08-12 23:40:43 +08:00
shenlan
4a113f42a6 Merge branch 'main' into nk5qth-codex/optimize-cli-server-with-langchaingo 2025-08-12 22:47:48 +08:00
shenlan
ecebcb6d64 docs: detail Milestone 2 subtasks 2025-08-12 22:32:45 +08:00
shenlan
915d7d8c3e Merge branch 'main' into codex/optimize-cli-server-with-langchaingo 2025-08-12 21:39:29 +08:00
shenlan
96e3c27b36 Update changelog.md 2025-08-12 21:34:46 +08:00
shenlan
04aab28a64 Merge branch 'main' into codex/optimize-cli-server-with-langchaingo 2025-08-12 21:27:56 +08:00
shenlan
224e6cdd23 docs: detail local and online models in tech stack 2025-08-12 21:25:00 +08:00
shenlan
325d17326b docs: add roadmap document 2025-08-12 19:33:52 +08:00
shenlan
e20b822fee docs: add embedding benchmark 2025-08-11 17:14:26 +08:00
shenlan
86f9de4496 docs: add benchmarking results 2025-08-11 17:01:52 +08:00
shenlan
27103f01c8 feat(api): allow custom askai timeout 2025-08-11 15:47:04 +08:00
Haitao Pan
e48c00eaba docs(pgvector): add common inspection SQL commands 2025-08-11 15:20:24 +08:00
Haitao Pan
6fe4e4d3d1 docs: move local embedding & Ollama API examples to api-endpoints.md 2025-08-11 09:03:05 +08:00
Haitao Pan
998ef1885b setup: make setup_macos_m4.sh idempotent and optimized for M4 Pro 2025-08-10 21:10:05 +08:00
Haitao Pan
048e3119fb optimize offline embedding server for macOS M4 Pro runtime 2025-08-10 21:00:21 +08:00
Haitao Pan
4b47863e6a merged: docs/offline_embed_server.py 2025-08-10 18:52:42 +08:00
shenlan
e1abcd4a1e Merge branch 'main' into codex/update-parameters-in-xcontrol-server-and-cli 2025-08-10 18:11:10 +08:00
Haitao Pan
5fd3078e3c add offline embedding server & model downloader 2025-08-10 17:38:45 +08:00
Haitao Pan
1c8876b69f docs: add offline embedding server & model downloader 2025-08-10 16:22:53 +08:00
Haitao Pan
98c3f12a62 docs: update /api/rag/upsert example to use upsert_1024.json for 1024-d test 2025-08-10 13:18:51 +08:00
shenlan
ed693a94aa Merge pull request #131 from svc-design/codex/add-embedding-configuration-fields
feat: add embedding config for rag
2025-08-10 10:41:44 +08:00
shenlan
aa373c9c23 feat: add embedding config for rag 2025-08-10 10:30:06 +08:00
shenlan
6f1e81687e Merge pull request #130 from svc-design/codex/replace-vector1536-with-vector1024
docs: update embedding dimension to 1024
2025-08-10 10:29:15 +08:00
shenlan
55b639c370 docs: update embedding dimension to 1024 2025-08-10 10:27:30 +08:00
shenlan
6fb9abf00f docs: add RAG upsert example and improve error handling 2025-08-10 10:18:35 +08:00
Haitao Pan
83a01546c2 docs: add hotfix.sql for schema fixes and make init.sql idempotent 2025-08-10 09:29:40 +08:00
shenlan
03680bc715 feat: ingest files automatically 2025-08-10 06:45:43 +08:00
shenlan
fb4fda84a8 docs: add pgvector DB initialization guide 2025-08-09 22:59:56 +08:00
shenlan
7377393848 feat: add rag upsert endpoint and retry cli 2025-08-09 22:28:25 +08:00
shenlan
6fb7a2e829 test: add default server config for tests 2025-08-09 21:42:44 +08:00
shenlan
5902ab8523 refactor: derive Chutes endpoint from provider config 2025-08-09 19:38:55 +08:00
shenlan
4c7082b7fd refactor: reorganize rag modules 2025-08-09 18:42:19 +08:00
shenlan
012f250412 feat: stream rag sync progress 2025-08-09 14:55:22 +08:00
shenlan
27054341bd Use default Redis port 2025-08-09 14:24:31 +08:00
shenlan
8117c89402 docs: clarify curl usage for rag query 2025-08-09 13:47:19 +08:00