OpenRAG is a comprehensive, single package Retrieval-Augmented Generation platform built on Langflow, Docling, and Opensearch.
Python
Commits on main
feat: route Azure AI Foundry over its OpenAI-compatible v1 surface (#2445)
2c1e56b Rico Furtado committed 1 month ago
fix: an upload can no longer replace a shared document without the permission to delete one (#2479)
94c0ea5 Rico Furtado committed 1 month ago
fix: reduce flakiness of tests and fix failing e2e tests (#2481)
cb7213d Lucas Oliveira committed 1 month ago
fix: a status with no entry no longer takes down the knowledge view (#2468)
f810b78 Rico Furtado committed 1 month ago
fix: embed through the gateway for every provider, not just aliased ones (#2443)
4872034 Rico Furtado committed 1 month ago
fix: pass account name explicitly for Azure Blob account key auth (#2454) (#2455)
fa6c13b Nathan Bakare committed 1 month ago
fix: make a duplicate check answerable from the server logs (#2461)
367fe31 Rico Furtado committed 1 month ago
fix: surface a failed duplicate check instead of ingesting anyway (#2434)
494bd4a Rico Furtado committed 1 month ago
fix: exclude both luna generations and their dated snapshots (#2444)
e036ecc Rico Furtado committed 1 month ago
feat: Improve RHOAI model provider error handling and resiliency (#2419)
fa2619d Mike Pawlowski committed 1 month ago
fix: keep IME confirmation Enter from sending the chat On macOS, Safari, Chrome, and Edge fire Enter when a Japanese candidate is confirmed. That key was treated as send, so the message went out before the user finished typing. (#2450)
5d28a2e Wallgau committed 1 month ago
fix: validate chunk settings before individual-file ingest (#2433)
4c1624f Nathan Bakare committed 1 month ago
fix: live chunk-settings validation and stop clobbering unsaved edits (#2430)
b632faf Nathan Bakare committed 1 month ago
fix: two files heading for one filename no longer merge into one document (#2406)
6bce2f8 Rico Furtado committed 1 month ago
fix: detect cloud-connector duplicates ingested from other sources (#2404)
9aa4ace Rico Furtado committed 1 month ago
fix: isolate watsonx.ai on-prem configuration and ingestion (#2422)
7a8b90d Edwin Jose committed 1 month ago
fix: require explicit Azure model selection on fallback (#2384) [forward-port to main] (#2420)
9606975 Edwin Jose committed 1 month ago
fix: return 5xx instead of unhandled error when index-admin service token is missing (#2382)
e238025 Nathan Bakare committed 1 month ago