Stepan Lebedev
600ff1ede0
feat(Unplug): chatting without brain streaming ( #970 )
...
* feat(Unplug): Adds new basic headless llm
* feat(Unplug): adds chatting without brain option when no streaming
* feat(Unplug): adds chatting without brain option when streaming
2023-08-18 10:32:22 +02:00
Stan Girard
7281fd905a
feat(microservices): split into 4 quivr to better handle long services ( #972 )
...
* feat(verbose): removed
* feat(microservices): split quivr into micro services
* test(main): fixed
2023-08-18 10:18:29 +02:00
joelsprunger
59ddfb4882
fix(chat routes): use brain model, temp, and token ( #902 )
2023-08-11 20:44:48 +02:00
Mamadou DICKO
6e777327aa
feat(chatMessages): add brain_id and prompt_id columns ( #912 )
...
* feat: add prompt_id and brain_id to chat history)
* feat: add prompt_id and brain_id to chat routes
2023-08-10 10:25:08 +02:00
ChloeMouret
e9ebeef72a
feat: gpt4 is not available for brains if there is no given openAiKey ( #850 )
...
* rename defineMaxToken
* use gpt-3.5-turbo instead of gpt-3.5-turbo-0613
* gpt4 not available if no open ai key
2023-08-07 16:35:23 +02:00
ChloeMouret
711e9fb8c9
refactor: delete common_dependencies function ( #843 )
...
* use function for get_documents_vector_store
* use function for get_embeddings
* use function for get_supabase_client
* use function for get_supabase_db
* delete lasts common_dependencies
2023-08-03 20:24:42 +02:00
Laurens Van Keer
68f03b2416
fix(chat routes): HTTPException import correction ( #833 )
...
* fix(chat routes): HTTPException import correction
Solves "TypeError: HTTPException() takes no keyword arguments"
* fix(chat routes): error in user id comparison in metadata endpoint
current_user.id is a UUID and must be converted to a string before comparison with the chat.user_id string
2023-08-03 12:11:25 +02:00
Mamadou DICKO
4ca6c667da
feat(backend): implement brain-prompt link ( #831 )
...
* feat: add prompt_id field to brain
* feat(Prompt controller): update prompt routes
* feat: remove unused private prompts
* refactor: add BrainEntity and repo and service
* tests: partially type main Repository
* feat: add PromptStatusEnum enum
* feat: change delete prompt repository return type
2023-08-03 10:37:13 +02:00
Noe
303ba72028
feat: Introduce repository pattern to prepare adding other database providers ( #646 )
...
* add sqlalchemy models
* add neon settings
* add insert brain
* abstract supabase from Brain class
* abstract supabase from Brain class
* abstract supabase from /models
* update Database to Repository
* update neon_tables to pg_tables
* update chat, api-key and message
* update vector class
* update settings
* update env vars for test
* Update backend-tests.yml
* fix test
* fix fetch_user_requests_count()
* fix fetch_user_requests_count()
* fix increment_user_request_count
* fix increment_user_request_count
* fix asset upload_response message
* fix pyright
* fix brain_subscription
* fix brain_subscription
* fix brain_subscription
* fix get user request stat
* update create_brain_user
* add delete brain vector and user
* add delete brain vector and user
* correctly call function
---------
Co-authored-by: Noé Pion <noe.pion@onfido.com>
Co-authored-by: raoufchebri <raouf@chebri.com>
Co-authored-by: Stan Girard <girard.stanislas@gmail.com>
2023-08-01 23:03:47 +02:00
Mamadou DICKO
edcbb30e97
fix: bugs ( #818 )
...
* feat: add chat config modal
* feat: save chat config in localStorage
* feat: remove <ConfigPage/>
* feat: overwrite chat config with brain
* fix(SettingsPage): upload payload keys
* fix: update default brain marker logic
* feat: set new created brain as current selected brain
2023-08-01 16:25:02 +02:00
Stan Girard
3166d089ee
feat(chat): added streaming ( #808 )
...
* feat(tmp): added streaming
* feat(streaming): implemented by changing order
2023-07-31 21:34:34 +02:00
Stan Girard
db40f3cccd
feat(llm): removing all llms to prepare for genoss ( #804 )
...
* feat(routes): removed all except openai
* feat(deadcode): removed some deadcode and summarization feature that wasn't used
* feat(streaming): removed privateGPT from it
* chore(requirements): increased version
2023-07-31 16:01:34 +02:00
Braden Mars
914689957d
fix(backend): FastAPI docs broken due to missing schema for NullableUUID ( #739 )
...
Just updates `NullableUUID` to extend from `UUID`. This lets FastAPI infer the correct field schema and resolves the /docs and /redocs endpoint failing to render
Signed-off-by: Braden Mars <bradenmars@bradenmars.me>
2023-07-23 11:14:42 +02:00
Matt
e61f437ce8
Feat/backend core ( #656 )
2023-07-17 07:57:27 +01:00
