Merged
Conversation
NorbertKlockiewicz
requested changes
May 7, 2025
| Alternatively, you can use `runInference`. It provides direct access to the model, without any wrapper, so the input string is passed straight into the model. If you're not sure what are implications of that, you're better off with `sendMessage` | ||
| Alternatively, you can use `generate` method. It allows you to simply pass chat messages and receive completion from the model. It doesn't provide any message history management. | ||
|
|
||
| If you need raw model, without any wrappers, you can use `forward`. It provides direct access to the model, so the input string is passed straight into the model. It may be useful to work with models that aren't finetuned for chat completions. If you're not sure what are implications of that, you're better off with `sendMessage` |
Contributor
There was a problem hiding this comment.
Suggested change
| If you need raw model, without any wrappers, you can use `forward`. It provides direct access to the model, so the input string is passed straight into the model. It may be useful to work with models that aren't finetuned for chat completions. If you're not sure what are implications of that, you're better off with `sendMessage` | |
| If you need raw model, without any wrappers, you can use `forward`. It provides direct access to the model, so the input string is passed straight into the model. It may be useful to work with models that aren't finetuned for chat completions. If you're not sure what are implications of that, you're better off with `sendMessage`. |
Collaborator
There was a problem hiding this comment.
I think you can mention here that you need to add special tokens to make it work
| import { View, StyleSheet, Text } from 'react-native'; | ||
|
|
||
| enum ModelType { | ||
| enum ModeType { |
Contributor
There was a problem hiding this comment.
Maybe just Mode instead of ModeType
| this.responseCallback(''); | ||
| this.isGeneratingCallback(true); | ||
| await this.nativeModule.runInference(input); | ||
| console.log('INPUT:', input); |
Contributor
There was a problem hiding this comment.
Remove console.log
Contributor
|
Can you also bump huggingface/jinja to newest version(0.5.0), as the current has problems with qwen 3 tokenizer. |
chmjkb
requested changes
May 7, 2025
Collaborator
There was a problem hiding this comment.
I believe that this is wrong, you're supposed to pass all the special tokens to forward(), but the above prompt definition doesn't do that. Change the call of this method to generate() (not sure if this one is ok)
| Alternatively, you can use `runInference`. It provides direct access to the model, without any wrapper, so the input string is passed straight into the model. If you're not sure what are implications of that, you're better off with `sendMessage` | ||
| Alternatively, you can use `generate` method. It allows you to simply pass chat messages and receive completion from the model. It doesn't provide any message history management. | ||
|
|
||
| If you need raw model, without any wrappers, you can use `forward`. It provides direct access to the model, so the input string is passed straight into the model. It may be useful to work with models that aren't finetuned for chat completions. If you're not sure what are implications of that, you're better off with `sendMessage` |
Collaborator
There was a problem hiding this comment.
I think you can mention here that you need to add special tokens to make it work
chmjkb
requested changes
May 7, 2025
| messages: Message[], | ||
| tokenizerConfig: any, | ||
| tools?: LLMTool[], | ||
| template_flags?: Object |
Collaborator
There was a problem hiding this comment.
Can you make template_flags camel case?
Contributor
Author
There was a problem hiding this comment.
Sorry, missed it, maybe we should enforce camel case with eslint? Can you create issue for it?
NorbertKlockiewicz
approved these changes
May 8, 2025
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Added generate function, and some little re…formats
Type of change
Tested on
Related issues
#226
Checklist