-
Notifications
You must be signed in to change notification settings - Fork 2
Add support for Anthropic models #5
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
packages/SemanticText.package/AnthropicConversationModel.class/instance/maxCompletionTokens.st
Outdated
Show resolved
Hide resolved
resolvedName | ||
|
||
self flag: #modelConstants. | ||
"Anthropic define '-latest' models directly, for which we already have definitions. No need to resolve anything." |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yeah, OpenAI defines these directly, too. It was meant to duplicate centsForCompletionToken
et al. so you don't have to repeat the numbers for the latest pinned model twice. Should probably explain this in a comment ^^
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ah, I understand now. Will fix that!
… in the Anthropic classes instead of asOpenAIObject; Remove rate limit data from Anthropic models and add a note that they're unimplemented; Make `getAnswers: number` with a number > 1 return an array of completions instead of throwing an error (I am not sure if number=1 should return an array of size 1 or just the completion itself).
…/instance/maxCompletionTokens.st Co-authored-by: Christoph Thiede <[email protected]>
…more closely what the Anthropic API sends us, handle more streaming events in preparation for streaming tool calls. Fix rawOutput tracking for Anthropic models. Fix an issue with getting multiple assistant replies.
(wip -- tool use support has not been added yet)