Google Vertex API support (#950)

[Trending assistants] make ui visible (#966)

[Assistants] trending feature (#938)

Expose sampling controls in assistants (#955) (#959)

Limit number of fetched conversations in the API (#954)

Dynamic system instructions (#949)

[Websearch] change context schema (#944)

🚸 When a user signs up they're directly logged in (#952)

add api assistants endpoints (#951)

Count system prompt tokens (#850)

Enhance Dynamic User Attribute Handling in OIDC Integration (#885)

feat: smooth and combine token output (#936)

Modals: dont close if text is selected (#943)

Update AssistantSettings.svelte (#941)

Store user agent in conversation (#940)

Prepend domain filters in search query generation (#932)

Fix model identification when creating conversations (#820)

Show error when webpage cannot be reached or parsed (#930)

Fix multi domains for assistants (#929)

Remove extra > in URLs (#931)

Anthropic Endpoint Support (#923)

Add websearch controls for assistants (#812)

Fix prompt caching on llama.cpp endpoints (#920)

fix: Prompt bypasses login (#921)

Add openai embeddings (#915)

Remove double meta tags on models page (#918)

Direct model page (#858)

Return an error on unknown model in new conversation (#916)

Make assistant settings form fill the modal (#900)

Include last chunk in websearch context (#912)

Make sure preprompt is set on open ai endpoint type (#913)

[Assisntans filter] More UX fixes (#908)

[Assistants filter] Make UI more responsive (#906)

Add limits on API endpoints (#886)

[Assistants filter] Fix bug (#902)

[Assistants] Show/enable UI for filtering (#901)

Automated migration setup (#897)

[Assistants] Filter on names (#841)

Fix issue with "continue" feature on llama.cpp endpoints (#898)

Bug Fix: Json Decoder aggessively pulls json (#867)

Navigation collapse (#893)

fix retry button showing while loading (#894)

Improve error handling when posting new messages (#892)

Make sure we properly close the stream when done (#891)

Fix Error: Ancestor Not Found issue (#884)

Use `uuid` for message ID generation instead of crypto method (#882)

Pass websearch to `preprocessMessages` (#876)

Small UX updates assistants (#872)

[Assistants] Fix `remove` behaviour (#869)

[Mongo] count messages (aggregate) only when needed V2 (#866)