You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
custom: 'Custom JSON parameters to send to the API. Must be valid JSON format.',
108
108
showThoughtInProgress: 'Expand thought process by default when generating messages.',
109
109
disableReasoningParsing:
110
-
'Send reasoning_format=none to prevent server-side extraction of reasoning tokens into separate field',
110
+
'Send reasoning_format=none so the server returns thinking tokens inline instead of extracting them into a separate field.',
111
111
excludeReasoningFromContext:
112
-
'Strip reasoning content from previous messages before sending to the model. When unchecked, reasoning is sent back via the reasoning_content field so the model can see its own chain-of-thought across turns.',
112
+
'Strip thinking from previous messages before sending. When off, thinking is sent back via the reasoning_content field so the model sees its own chain-of-thought across turns.',
113
113
showRawOutputSwitch:
114
114
'Show toggle button to display messages as plain text instead of Markdown-formatted content',
115
115
keepStatsVisible: 'Keep processing statistics visible after generation finishes.',
'Enable Python interpreter using Pyodide. Allows running Python code in markdown code blocks.',
147
147
preEncodeConversation:
148
-
'After each response, automatically re-submit the conversation (without chain-of-thought) to pre-fill the server KV cache. This makes the next turn faster since the prompt is already encoded while you read the response.',
148
+
'After each response, re-submit the conversation to pre-fill the server KV cache. Makes the next turn faster since the prompt is already encoded while you read the response.',
149
149
enableContinueGeneration:
150
150
'Enable "Continue" button for assistant messages. Currently works only with non-reasoning models.'
0 commit comments