pascalkienast 82328b843d Retry LLM payloads with reasoning token params after 400/422 ...

Some providers reject temperature/max_tokens. Remap once to
max_completion_tokens instead of matching model name strings.

2026-08-18 17:27:02 +00:00

..

AppInfo

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

Controller

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

Db

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

Exception

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

Jobs

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

Listener

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

Migration

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

Reference

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

Service

Retry LLM payloads with reasoning token params after 400/422

2026-08-18 17:27:02 +00:00

Settings

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

ToolProvider

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

Webhook

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00

TypeDefinitions.php

Initial open-source release of Talk AI

2026-07-08 21:13:13 +02:00