{"full_name":"neuml/txtai","owner":"neuml","name":"txtai","url":"https://github.com/neuml/txtai","description":"💡 All-in-one AI framework for semantic search, LLM orchestration and language model workflows","homepage":"https://neuml.github.io/txtai","topics":[],"language":"Python","license":"Apache-2.0","avatar":"https://avatars.githubusercontent.com/u/59890304?v=4","created_at":"2020-08-09T19:14:59Z","pushed_at":"2026-11-10T02:56:25Z","metrics":{"stars":13255,"forks":988,"watchers":113,"issues":1206,"prs":309,"releases":68,"issues_open":null,"issues_closed":null,"prs_open":null,"prs_closed":null,"prs_merged":null,"discussions":0},"growth":{"day":{"stars":65,"forks":7,"base_day":"2026-11-09"},"week":{"stars":198,"forks":30,"base_day":"2026-11-03"},"month":{"stars":257,"forks":61,"base_day":"2026-10-11"}},"history":{"dates":["2026-09-06","2026-09-07","2026-09-08","2026-09-09","2026-09-10","2026-09-11","2026-09-12","2026-09-13","2026-09-14","2026-09-15","2026-09-16","2026-09-17","2026-09-18","2026-09-19","2026-09-20","2026-09-21","2026-09-22","2026-09-23","2026-09-24","2026-09-25","2026-09-26","2026-09-27","2026-09-28","2026-09-29","2026-09-30","2026-10-01","2026-10-02","2026-10-03","2026-10-04","2026-10-05","2026-10-06","2026-10-07","2026-10-08","2026-10-09","2026-10-10","2026-10-11","2026-10-12","2026-10-13","2026-10-14","2026-10-15","2026-10-16","2026-10-17","2026-10-18","2026-10-19","2026-10-20","2026-10-21","2026-10-22","2026-10-23","2026-10-24","2026-10-25","2026-10-26","2026-10-27","2026-10-28","2026-10-29","2026-10-30","2026-10-31","2026-11-01","2026-11-02","2026-11-03","2026-11-04","2026-11-09","2026-11-10"],"stars":[12927,12927,12928,12931,12935,12940,12943,12944,12946,12947,12947,12948,12951,12956,12962,12964,12964,12966,12968,12971,12977,12981,12985,12985,12985,12988,12989,12991,12990,12991,12991,12990,12991,12995,13001,12998,13004,13006,13011,13014,13018,13017,13023,13023,13024,13028,13028,13027,13027,13029,13032,13034,13038,13041,13043,13045,13049,13053,13057,13061,13190,13255],"forks":[889,889,890,889,890,890,891,891,890,890,890,890,891,891,890,891,892,894,896,898,899,901,900,901,903,906,908,909,911,914,915,917,920,921,923,927,927,930,931,932,945,953,992,1027,1077,1119,1089,945,947,947,951,951,952,953,954,956,956,956,958,963,981,988],"watchers":[115,115,116,116,116,116,116,116,116,116,116,116,116,116,116,116,116,116,116,116,114,114,114,114,114,114,113,113,113,113,113,112,112,112,112,112,112,113,113,113,113,114,114,114,114,113,113,113,113,113,111,111,111,111,111,111,111,111,111,111,113,113],"issues_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null],"prs_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null]},"boards":{},"same_owner":[],"issues":[{"number":1161,"title":"`mcpadapt` not working with latest version of `mcp`","comments":8,"created_at":"2026-07-28T15:00:12Z"},{"number":1517,"title":"Add a persisted multivector token store for exact MaxSim","comments":4,"created_at":"2026-11-05T01:30:03Z"},{"number":1536,"title":"Batched Task result packing gets quadratically slower with the batch size","comments":0,"created_at":"2026-11-08T18:33:22Z"},{"number":1534,"title":"SQL query fails when a negative number follows an operator without a space","comments":0,"created_at":"2026-11-08T16:27:37Z"},{"number":1532,"title":"Embeddings search misses Cypher queries with lowercase keywords or non-space whitespace","comments":0,"created_at":"2026-11-08T15:09:06Z"},{"number":1530,"title":"Keyword and IVFSparse search fail when CPU count is undetermined","comments":0,"created_at":"2026-11-08T03:07:39Z"},{"number":1528,"title":"Workflow task iterator outputs fail during result packing","comments":0,"created_at":"2026-11-07T10:35:32Z"},{"number":1518,"title":"Log-odds calibration uses upper middle score instead of median for even candidate sets","comments":0,"created_at":"2026-11-05T11:18:27Z"}],"discussions":[],"latest_release":{"name":"v9.14.0","tag":"v9.14.0","published_at":"2026-10-06T12:49:05Z","body":"**This release adds new ANN backends for RaBitQ and Zvec Sparse, native support for ONNX vectors along with a bonanza of improvements and bug fixes**\r\n\r\nSee below for full details on the new features, improvements and bug fixes.\r\n\r\nNew Features\r\n--------------------------\r\n\r\n- Add ONNX vectors backend (#1211, #1212) @joseph-bergin\r\n- Add RaBitQ as an ANN backend (#1244, #1251) @pma1999\r\n- Add Zvec Sparse ANN (#1280, #1307) @Ni0ri\r\n- Require Python 3.11 (#1273)\r\n\r\nImprovements\r\n--------------------------\r\n\r\n- Update vectors package to standardize fallback logic (#1221)\r\n- Support pooling_mode with models trained with recent Sentence Transformers (#1292)\r\n- Refactor ANN unit tests (#1305)\r\n- Support trust_remote_code in associated AutoModels (#1342)\r\n\r\nBug Fixes\r\n--------------------------\r\n\r\n- Latest version of LiteLLM is broken with Python 3.10+ (#1193)\r\n- Fix bash tool hanging on commands that read stdin (#1209, #1210) @pacocartones\r\n- fix(api): parse the graph query parameter as a bool (#1213) @serhiizghama\r\n- Load the Milvus collection when opening a saved index (#1214) @morgan-coded\r\n- Ignore padding rows in LateEncoder scores (#1215) @morgan-coded\r\n- Resolve label filters by label id (#1216) @morgan-coded\r\n- Improve late pooling configuration errors (#1217) @morgan-coded\r\n- Keep count() accurate after repeated deletes in GGML and IVFSparse (#1218) @morgan-coded\r\n- Delete every index entry for an id in Terms (#1219) @morgan-coded\r\n- fix(vectors): remove the sparse spool file when checkpointing is disabled (#1220) @serhiizghama\r\n- Keep the archive working directory alive when saving a loaded index (#1222) @morgan-coded\r\n- Key the vectors model cache by configuration (#1223) @morgan-coded\r\n- Save empty keyword and content indexes without errors (#1224) @morgan-coded\r\n- fix(workflow): add request timeout to ServiceTask (#1225, #1226) @pacocartones\r\n- Select the top documents in LateEncoder by the document axis (#1227) @morgan-coded\r\n- Fix ANN delete on quantized Torch stores and tiny GGML indexes (#1228) @morgan-coded\r\n- Fix Transform.stream() yielding a None text value into the encoder (#1231) @AmirF194\r\n- fix(ann): honor quantize: false in the Faiss and SQLite backends (#1234) @serhiizghama\r\n- Fix quantize: false silently enabling Q8_0 quantization in the GGML backend (#1235, #1236) @AmirF194\r\n- Fix Cluster combining avg() as an unweighted mean of shard averages (#1237) @AmirF194\r\n- fix(database): don't drop rows when sorting or grouping by a JSON field (#1238) @serhiizghama\r\n- Latest version of Transformers has breaking change for Text to Audio pipeline (#1239)\r\n- Save IVFSparse indexes after deleting with NumPy ids (#1241) @morgan-coded\r\n- Set the HNSW delete counter when loading an index (#1242) @morgan-coded\r\n- Keep community nodes that match none of the topic terms (#1243) @linhongyu510\r\n- fix(api): forward zero weights and url-encode parameters in Cluster search (#1245) @serhiizghama\r\n- Honor the nfeatures setting in IVFSparse (#1246, #1250) @liugj0710\r\n- Keep deleted keyword results excluded after common-term scoring (#1247, #1248) @LimbC-C\r\n- Keep SQLite ANN data when saving a modified index to a new path (#1249) @Irahan2\r\n- Fix DuckDB.formatargs dropping repeated named bind parameters (#1252, #1253) @AmirF194\r\n- Preserve local filenames in RetrieveTask (#1255, #1256) @dingpuyu\r\n- Support explicit LIKE escapes in wildcard term queries (#1258, #1259) @LimbC-C\r\n- Preserve falsy TemplateTask rule matches (#1261, #1262) @HuaTNA\r\n- Label QA spans without the answer with the CLS token (#1263, #1264) @huiq777\r\n- Fix training with pandas DataFrames (#1265, #1266) @huiq777\r\n- Preserve reusable RagTask inputs (#1267, #1268) @HuaTNA\r\n- Pack rows into chunks up to and including maxlength (#1269, #1270) @huiq777\r\n- Reject iterator inputs for multi-action tasks without buffering (#1271, #1272) @LimbC-C\r\n- Preserve keyword document ID types across save/load (#1274, #1275) @HuaTNA\r\n- Fall back to root tokenizers for ONNX and LiteRT"},"readme":true,"readme_zh":false,"news":[{"title":"Prompt-driven vector search with LLMs","url":"https://github.com/neuml/txtai","hn":"https://news.ycombinator.com/item?id=34633043","source":"Hacker News","lang":"en","date":"2023-02-02","points":28,"comments":5},{"title":"Show HN: Txtai – AI-powered search engine in JavaScript","url":"https://github.com/neuml/txtai.js","hn":"https://news.ycombinator.com/item?id=25223811","source":"Hacker News","lang":"en","date":"2020-11-26","points":6,"comments":0},{"title":"Txtai 5.0 released – build semantic search applications and workflows","url":"https://github.com/neuml/txtai","hn":"https://news.ycombinator.com/item?id=32998298","source":"Hacker News","lang":"en","date":"2022-09-27","points":5,"comments":1},{"title":"Show HN: Txtai 5.2 released: open-source semantic search","url":"https://github.com/neuml/txtai","hn":"https://news.ycombinator.com/item?id=34079703","source":"Hacker News","lang":"en","date":"2022-12-21","points":4,"comments":0},{"title":"Txtai 5.1: better machine translation, Whisper transcription and ARM support","url":"https://github.com/neuml/txtai","hn":"https://news.ycombinator.com/item?id=33261773","source":"Hacker News","lang":"en","date":"2022-10-19","points":4,"comments":1},{"title":"txtai: A FOSS Vector ML Database for Intelligent Semantic Search","url":"https://github.com/neuml/txtai","hn":"https://news.ycombinator.com/item?id=32918254","source":"Hacker News","lang":"en","date":"2022-09-20","points":4,"comments":0},{"title":"Show HN: The Semantic Graph","url":"https://github.com/neuml/txtai/blob/master/examples/38_Introducing_the_Semantic_Graph.ipynb","hn":"https://news.ycombinator.com/item?id=32870191","source":"Hacker News","lang":"en","date":"2022-09-16","points":3,"comments":1},{"title":"Show HN: Txtai – SQL-driven semantic search with machine learning functions","url":"https://github.com/neuml/txtai","hn":"https://news.ycombinator.com/item?id=30900702","source":"Hacker News","lang":"en","date":"2022-04-03","points":3,"comments":1},{"title":"txtai：開源框架助你輕鬆建構語義搜尋與 LLM 工作流程","url":"https://www.techritual.com/2026/08/14/529172/","source":"techritual","lang":"zh","date":"2026-08-14"},{"title":"txtai：一站式 AI 框架 解鎖語義搜尋與 LLM 工作流","url":"https://www.techritual.com/2026/08/13/529166/","source":"techritual","lang":"zh","date":"2026-08-12"},{"title":"【GitHub项目推荐--txtai：全功能开源AI框架】","url":"https://blog.csdn.net/j8267643/article/details/156492164","source":"blog.csdn","lang":"zh","date":"2026-07-11"},{"title":"Discover the AI Breakthroughs of 2023: Dolma, StableVideo, OpenPipe, txtai, and Project IDX","url":"https://lablab.ai/ai-articles/the-ai-breakthroughs-of-2023","source":"lablab","lang":"en","date":"2023-09-20"},{"title":"Complete Tutorial On Txtai: An AI-Powered Search Engine","url":"https://analyticsindiamag.com/deep-tech/complete-tutorial-on-txtai-an-ai-powered-search-engine","source":"Analytics India Magazine","lang":"en","date":"2020-11-09"},{"title":"掌握RAG系统的七个优秀GitHub存储库","url":"https://news.google.com/rss/articles/CBMiU0FVX3lxTE5DbUswTVdkaXYxak9sYTFVUE5ILTRwRXI0TU16cGtZbXMtYzJoUmhXeXUtZGZXdWpVZ1ZaZFdnbEszcWVVakZ2TEo2Rlo1MGNIUFJv?oc=5","source":"51CTO","lang":"zh","date":"2025-10-29"},{"title":"2025年值得关注的21个RAG开源项目","url":"https://news.google.com/rss/articles/CBMiakFVX3lxTFBHcHZNODhuWVFvdldka1hMVllhRlpGZEh0dFFzcUkzVVF0LU9KNEtQYnoyRFhpeGVCUWhCNDBZT0hlYzhreHFfVTAwU0diSXpiU3plQTVOTy1MUzBzZGZJYms5cHF1QkZjZXc?oc=5","source":"53AI","lang":"zh","date":"2025-04-12"},{"title":"2024 GitHub 十大最佳 RAG 框架","url":"https://news.google.com/rss/articles/CBMiXEFVX3lxTFB4Zk9iUFVQaWFhZnJBWHMydFJqbUp4dVBTYmgwR28tQUdtbkZKZVcxU2RsYVY3bXMxODVWRTE4MW9xaERiRWdzRFoyVUdUQ1VGTEZkOEtXOVdwN3o4?oc=5","source":"53AI","lang":"zh","date":"2025-01-23"},{"title":"Github上的十大RAG(信息检索增强生成)框架","url":"https://news.google.com/rss/articles/CBMiVkFVX3lxTFBBLXc5bmN2TDhYUC1yVExmNk1ld1pYQ1ItMDhHNjh6VjVQZl82ZUpkUF9OSEFZUE1vVG9ZN0poVVBJT0FSbldTZEtZdGx5ZGpBaEhmUUtR?oc=5","source":"QQ News","lang":"zh","date":"2024-10-31"}],"stale":false,"snapshot_at":"2026-11-10T09:35:25Z","history_start":"2026-09-06"}