Refactor logger error messages for consistency. (#1257)

Handle errors gracefully in building PlaywrightBlocker (#1248)

fix websearch on assistants (#1244)

Fix: tools in endpointVertex now optional (#1234)

Add sampling parameter support to cloudflare (#1233)

Fix error logging (#1228)

Trim trailing comma on tool calls (#1229)

Send `ollama pull` request if model is not available locally (#1227)

feat: mime types on tools, show upload btn on demand (#1204)

feat: skip tools when all disabled (#1205)

Add support for SearchApi API (#1196)

feat: request model be concise on tool image outputs (#1202)

Generic tool error msg "Error occurred" (#1193)

Simplify tool error handling (#1168)

Reduce message update size returned by websearch (#1179)

Log parsing failures, drop from UI (#1189)

Bypass spaces tools rate limit (#1167)

Fix Playwright leak and use multiple contexts (#1187)

Add metrics for models, tools, websearch (#1186)

fix: parse ISS if returned in OIDC flow (#1162)

Ony call tools if model.tools is true (#1172)

Always close playwright page (#1171)

Clear playwright singleton on close (#1170)

Add error message for tools (#1163)

Function calling (#996)

Generic Multimodal Support Fixed (#1147)

Revert "Generic Multimodal Support" (#1146)

Generic Multimodal Support (#1021)

Web Search: Playwright, spatial parsing, markdown (#1094)

[fix] Assistant websearch text/markdown source (#1129)

Extend endpointOai.ts to allow usage of extra sampling parameters (#1032)

feat: add support for anthropic on vertex (#958)

Fix context of logger (#1112)

Fix OpenAI Embeddings Auth Issue (#1077)

[VertexAI] Add support for tools parameter (#1065)

Move vars to dynamic, add metrics (#1085)

Use pino for logs (#1086)

set `sameSite` to `lax` when allowing insecure cookies (#1078)

Add `ALLOW_INSECURE_COOKIES` feature flag (#1076)

Use TEI space in prod (#1051)

Support Gemini 1.5 Pro from Vertex AI (#1041)

Always dispose of pipeline for embeddings after use (#1048)

Dispose of embeddings pipeline occasionally to clear memory (#1047)

Kill app on tokenizer load fail (#1044)

return empty if error on hfapi endpoint (#1038)

Use inference API for embeddings in huggingchat prod (#1037)

Switch task model and add max tokens limits (#1036)

Add logs to tokenizer fetch (#1030)

Add langserve endpoint (#1009)

Move default template so it doesn't override tokenizer (#987)