v1.0.0

1.0.0 (2024-10-28)

Bug Fixes

• Change organisation into teams (#132) (235ca88)
• A better landing page (#23) (9199efa)
• Ability to call external API's (#45) (39825fd)
• Add Airbyte docs (5aa08fa)
• Add bionic command (#419) (6023234)
• Add certs (0915aca)
• Add copy button for responses (#535) (473a329)
• Add CORS headers to API. (4438ce3)
• Add new images (97a2cdc)
• Add Prometheus Metrics (#514) (ea2b2b5)
• Add RabbitMQ connector (#386) (ef45eaf)
• Add release (328f4f7)
• Add step (8eb97da)
• Add step attribute to temperature form field (#93) (95e3a1f)
• Add stop button (#629) (f4d21a9)
• Add testing data (c9ba63b)
• Airbyte to Dataset (#387) (34c647c)
• Align with Rust on Nails architecture (#447) (3375da8)
• Allow configuration of chunking (#48) (516480d)
• Allow multiple document upload (af538e8)
• Allow non-streaming chat completion API calls. (#605) (354c74d)
• Allow user to set trim ratio (#159) (c0a4233)
• Another attempt to fix tests. (76185ce)
• Another Integration Attempt (939c758)
• API to create prompts based on dataset. (#40) (cc6807a)
• Apply namespace (289352e)
• Assistants should have categories (#640) (c0c8a3f)
• Attach Keycloak to Postgres (#414) (32c3bd9)
• Auto deployments to AWS (#467) (0372578)
• Bare metal (#409) (ccb7ba6)
• Better error handling (4a110b4)
• Build envoy container (8b73d7a)
• Build error (d4561bd)
• Build error (581fb8b)
• Build keycloak container (1d7c652)
• Build keycloak container (5d7e73a)
• Caps for SSE (4fca4ca)
• Cascade deletes to chunks from datasets (#394) (57315d3)
• Change name (ee96d1b)
• Change ownership (855f4ed)
• Changed operator so that it does a server side apply rather than create for resources. (#470) (99fba24)
• Check authz for API keys. (4dc40cb)
• Choose correct file (9fc3545)
• Configurable logout utl (#283) (e033bc5)
• Configure access to the chunking engine (aebf143)
• Connect prompts to the models. Use the prompts in the console. (#18) (a00bae7)
• Console on firefox (29a0c78)
• Contenxt size (ec19e48)
• Context size of Llama 2 7B is 2K (aabc47d)
• Correct Import (9356c31)
• Correctly address dashboard (2569968)
• Correctly set name (f51779a)
• Create a folder for dashboards (#625) (49af409)
• Create the operator (#300) (548490b)
• Dataset Privacy (#481) (93dc749)
• Default embeddings model (#81) (ab541cb)
• Delete members or invites from a team. (#616) (17541ed)
• Deploy Oauth2 Proxy and Envoy. (#309) (2322a2f)
• Develop on Kubernetes (#426) (4f76ca0)
• Dioxus update (#430) (f4be4c2)
• Disable delete chat via rbac (#577) (37a918b)
• Disconnect models and AI Assistants. (#553) (fc57298)
• Docs (5029dd5)
• Document deployment (#140) (f056029)
• Don't send max_tokens (0501e43)
• Don't split streaming JSON. (#361) (4775386)
• Don't use a node port for grafana (ca5327d)
• Drawer background (b47b995)
• Encapsulate models (#331) (246485d)
• Enhanced AI Assistants (#559) (e2f5850)
• Ensure rag engine is dpeloyed (362dbf3)
• expected string, got val (#526) (ddfd612)
• Export Linux (a5836ef)
• Export video as an asset (0c385e8)
• Fix issue after registration where we need to set a category for the system prompt. (92450e7)
• Fix issue with Company prompt (#485) (af0f3c6)
• Fix migrations (#22) (2b6f87f)
• Fix tests (9882f8f)
• Fix tests (eb23e67)
• Fix uploads (#492) (79e550b)
• Fixed bug with where you couldn't delete a model (940d228)
• Fixed issue connecting promps to datasets. (#441) (2a3f13d)
• Fixed Issues with TLS (#440) (12b5ef0)
• Flag for ingress (#497) (be7b422)
• Generate all secrets from the operator (#403) (db729fe)
• Generate cloudflare config (#518) (a2dfeea)
• Generate markdown on the fly (#64) (2d65bb9)
• Get integration testing into the CI/CD pipeline (#85) (108b1cc)
• Get pods (2b5f07d)
• Get the API working (#96) (3cc8a45)
• Global datasets (#583) (610d81c)
• Global datasets should get added to global assistants (#609) (20660c1)
• Goat counter (#548) (bef4361)
• GPU setting for operator (#400) (54fec26)
• Guide users towards kubernetes (#381) (a997110)
• Hash issue (3097e87)
• Have a central location to handle object storage (#654) (1053b3d)
• History table (#636) (3764c55)
• Identify and attempt in place upgrade (#528) (4fdaad0)
• Ignore localstorage when value not present (784b119)
• Incorprate Ingress Controller (310d8a7)
• Increase upload size and add docs (f6cbc38)
• Install Models K8's (#337) (5246af6)
• Install nginx by default (6281cc9)
• Installation documentation (#26) (dcae556)
• Integration testing (433cf61)
• Integration Testing (427c6ca)
• Integration testing (#269) (bd0c3aa)
• Integration Testing (#276) (8d8e59c)
• Integration Testing (#439) (c3ec6c5)
• Invitations now work (e4d945d)
• Issue accessing API. (#501) (a180141)
• Issue adding multiple documents (#443) (995b109)
• K3s running gen AI at the edge. (#405) (fb418e5)
• K8's Operator In Progress (#299) (58ed1d9)
• K8s Operator (#294) (f5d56b5)
• K8s operator (#305) (2bdd59c)
• Keycloak on /oidc (9edaec0)
• Kubernetes (#412) (b55180f)
• Limits form now works (#468) (424901c)
• Linux musl (#424) (ead6d12)
• Listen on the correct port (e525109)
• Llama 7b chat (#72) (5b4691b)
• LLama3 8B (ee734cd)
• Load the correct model into prompt (7946c7a)
• Load webpki roots (b19d0ee)
• Mailhog was blocking Multi tenancy (#632) (c1ddc11)
• Make it easier for users to do Proofs of concept. (#484) (0890190)
• Make pipeline job more robust (#151) (9c0e771)
• Max upload size and API auth. (#180) (e95d199)
• Migrations (d3bfc8d)
• Migrations should run without role permissions (#150) (ff089e5)
• Model dialog and RBAC (#657) (eed3d9a)
• Model selector (again) (#639) (d91b5de)
• Modl duplicates (42914c1)
• More debug on installer (8edf093)
• More robust handling of newlines for chat completion LLM API (#30) (fde80ca)
• More testing (bb3e80e)
• More typed routes (#449) (9c30628)
• Moved config files (b4d751c)
• Narrow down teams (#487) (eae3545)
• New embeddings end point (f91e113)
• Numerous UI enhancements (#634) (448efba)
• ollama docker-compose (#169) (8fb3f04)
• Only set role if we have permission (d95a848)
• Option to disable ingress (396fc13)
• Pass in hostname url (#429) (37f44eb)
• Patch openai (1e0fade)
• Patch openai node package to work with TGI (#384) (7d55380)
• pgAdmin and GPU (#402) (f81df29)
• PgAdmin via ingress (#413) (5d03891)
• Pipeline job in Kubernetes (#312) (e6515aa)
• Point to correct pod (55a9115)
• Pre-load models into embeddings engine (#335) (6060c7d)
• Prep for Observability (#624) (ed8dfc1)
• Preperation for new Assistants UI (#649) (459f3a8)
• Privacy (#483) (d593dd7)
• Push bionic-rag-engine image (707cea9)
• Push correct images (20fd178)
• Put keycloak behind a proxy (4172319)
• Refactor Operator (#422) (81dd17d)
• Refactor pages and remove code. (#115) (f56149b)
• Reference correct param (0c124b4)
• Reference in a different way (65bd0dc)
• Refresh the page during document upload (#379) (d772107)
• Release (51d90cd)
• Remove keycloak from operator (#362) (82f9a56)
• Remove stream timeout which was cuting off inference after 25 seconds. (#466) (fdfcf49)
• Remove upgrade option (9eab60f)
• Restart pods after upgrade (9d0a27b)
• Return Auth error and add API to integration tests (#311) (bcc37b6)
• Return corect error code (f564ead)
• Revert (bdc81b9)
• Run tests after build (36cf55a)
• Saas issues (#524) (2273b50)
• Save streamed chat results server side. (#462) (ebcf0bd)
• Search chat history (#537) (a9f4c8f)
• Security check (c6fe0d1)
• Security fix (#398) (3bd3434)
• Set assistant permissions (#572) (8b0c0ad)
• Set correct port in K8's (37c9a9a)
• Set created by (d401962)
• Set to cluster IP (cac3bcb)
• Setup RBAC permissions (#198) (075042c)
• Show address when finished (581c0db)
• Show enterprise features (#521) (6948ba3)
• Show images against assistants. (#651) (43a4585)
• Show Pods (ea60825)
• Show top usage users (#129) (5c81113)
• SMTP now configured via a kubernetes secret (#479) (d98dae0)
• Split integration testing (205b0a8)
• Start to add integration testing (3e935e5)
• Start to add integration testing (54fde83)
• Start to add integration testing (f341dd2)
• Start to add integration testing (71f64ca)
• Start to add integration testing (dbff1aa)
• Start to integrate key cloak in kuberenetes (#306) (30f3b74)
• Static website PoC (#542) (cfd9380)
• Streaming (382fae3)
• Streaming in Bionic Local (#504) (ecac783)
• Switch to tailwind for utilities (#110) (6706e37)
• System Administrator (#189) (387f995)
• Team access (#284) (f90c32f)
• tests (4d849d6)
• Text overflow in assistant examples (87877a1)
• The K8s operator should be in it's own namespace (85dba90)
• Tighten up on context size (#156) (b73efd0)
• TLS mode to prefer (009b0fa)
• Token limits functionality (#512) (955934e)
• Track API chats (#452) (843bb62)
• Trailing space (1a7bc34)
• Truncate search results (c7ee0c9)
• Try build (bb62e3a)
• Try on path /oidc (e35cbc3)
• Typed routes (#448) (f89a10e)
• Ui ux misc (#56) (6d0f9d1)
• UI/UX issues with the console (#20) (88ee8db)
• Update PgAdmin (#411) (b29f4ee)
• Update version (21b1835)
• Update version (27b935e)
• Update versions (525c6cd)
• Update versions (df3f01b)
• Upgrade Local AI. (#33) (59cf202)
• Upgrade to latest Rust and fix warnings (2a4adb2)
• Use /chat/completions (#59) (3152ee7)
• Use correct table name (1351e50)
• Use correct version of inference engine (cf9b80a)
• Use cria as llama backend (3d07714)
• Use cria for completions (#67) (7f1c688)
• Use embeddings model attached to dataset. (#378) (83e17bf)
• Use github token (a56b5e6)
• Use github token (7a6503c)
• Use https when required (#99) (d693806)
• Use llama-2-7b-chat as the model. (#60) (710cb51)
• Use OpenAI API node library (#368) (ab9dafa)
• Use the correct model address (12ea6e1)
• Use the new chat API (#49) (4c82074)
• Various gui issues (#179) (78311fc)
• Version exe (38694e1)
• Versions (9f4e47c)
• We need the patches (a3c3ae4)

Features

• Add text to speech capability (#611) (28d6a25)
• Audit trail for system actions and usage (#123) (cb810bf)
• Chat history (#88) (a0e3928)
• Implement Document Pipelines (#101) (efafbe1)
• Internal API (#35) (cf72a15)
• Openid connect for Authn and Authz (#214) (0b37b01)
• Rate Limiting (#454) (346717e)
• Show the user the context used for a RAG query. (#371) (fff808d)