♻️ Simplify setup, prepare for multi-models #156
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Closed
julien-c
reviewed
May 3, 2023
Grsmto
reviewed
May 3, 2023
julien-c
requested changes
May 3, 2023
| @@ -6,6 +6,7 @@ import { | |||
| PUBLIC_DEPRECATED_GOOGLE_ANALYTICS_ID, | |||
| } from "$env/static/public"; | |||
| import { addYears } from "date-fns"; | |||
| import { inspect } from "node:util"; | |||
There was a problem hiding this comment.
Suggested change
| import { inspect } from "node:util"; |
| @@ -2,4 +2,6 @@ export interface Message { | |||
| from: "user" | "assistant"; | |||
| id: ReturnType<typeof crypto.randomUUID>; | |||
| content: string; | |||
| // Only for "assistant" messages | |||
| model?: string; | |||
There was a problem hiding this comment.
IMO this should instead be on Conversation (and always be defined, i.e. we can migrate current DB)
Otherwise how do you export the conv data to a model author?
Suggested change
| model?: string; |
There was a problem hiding this comment.
@gary149 wanted to be able to switch models during a conversation
Otherwise how do you export the conv data to a model author?
Since we send the whole exchange everytime to the model during the inference, it makes sense to send to the model author the whole conversation up to the last response by their own model
| @@ -30,10 +31,14 @@ export async function POST({ request, fetch, locals, params }) { | |||
| const json = await request.json(); | |||
| const { | |||
| inputs: newPrompt, | |||
| model, | |||
There was a problem hiding this comment.
or pass a conv id instead of a model here, and get the model from the DB
ice91
pushed a commit
to ice91/chat-ui
that referenced
this pull request
Oct 30, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In preparation to open-sourcing the repo, simply the config needed:
Incidentally, this refacto helps prepare for multi-model
Note: this loses the distinction betweenMODEL_NAMEandMODEL_IDfor the sake of configuration simplicity (and I think that's for the better, but can be reverted)(don't forget to update your .env.local to test with prod model if you want - otherwise just setting
HF_ACCESS_TOKENin.env.localis enough)With default configuration: