Skip to content

Fix max tokens and add model options #436

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.

Sign up for GitHub

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

Open
wants to merge 10 commits into
base: main
Choose a base branch
from
Open

Fix max tokens and add model options #436

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
0d927f8
fix max_tokens in request
RedwindA Aug 8, 2023
0a754c5
standard fix
RedwindA Aug 9, 2023
7081707
Update api.ts
RedwindA Aug 9, 2023
a7b5d6b
Update api.ts
RedwindA Aug 9, 2023
ca539a6
更新 api.ts
RedwindA Aug 10, 2023
ccc0496
add models
RedwindA Aug 10, 2023
0399dec
add models
RedwindA Aug 10, 2023
96e8670
Fixed the calculation method of max_tokens
RedwindA Aug 15, 2023
ecbb890
Update chat.ts
RedwindA Aug 15, 2023
b0630ee
fix max_tokens in request
RedwindA Aug 15, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 10 additions & 2 deletions src/api/api.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,8 @@
import { ShareGPTSubmitBodyInterface } from '@type/api';
import { ConfigInterface, MessageInterface } from '@type/chat';
import { isAzureEndpoint } from '@utils/api';
import { modelMaxToken } from '@constants/chat';
import countTokens from '@utils/messageUtils';

export const getChatCompletion = async (
endpoint: string,
Expand Down Expand Up @@ -32,13 +34,16 @@ export const getChatCompletion = async (
}
}

const tokenCount = countTokens(messages, config.model);
const maxTokens = Math.min(config.max_tokens, modelMaxToken[config.model] - tokenCount-1);

const response = await fetch(endpoint, {
method: 'POST',
headers,
body: JSON.stringify({
messages,
...config,
max_tokens: undefined,
max_tokens: maxTokens,
}),
});
if (!response.ok) throw new Error(await response.text());
Expand Down Expand Up @@ -77,13 +82,16 @@ export const getChatCompletionStream = async (
}
}

const tokenCount = countTokens(messages, config.model);
const maxTokens = Math.min(config.max_tokens, modelMaxToken[config.model] - tokenCount-1);

const response = await fetch(endpoint, {
method: 'POST',
headers,
body: JSON.stringify({
messages,
...config,
max_tokens: undefined,
max_tokens: maxTokens,
stream: true,
}),
});
Expand Down
16 changes: 12 additions & 4 deletions src/constants/chat.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,9 +22,10 @@ export const modelOptions: ModelOptions[] = [
'gpt-3.5-turbo-16k',
'gpt-4',
'gpt-4-32k',
// 'gpt-3.5-turbo-0301',
// 'gpt-4-0314',
// 'gpt-4-32k-0314',
'gpt-4-0314',
'gpt-4-32k-0314',
'gpt-3.5-turbo-0301',
'gpt-3.5-turbo-0613',
];

export const defaultModel = 'gpt-3.5-turbo';
Expand Down Expand Up @@ -100,7 +101,14 @@ export const _defaultChatConfig: ConfigInterface = {
top_p: 1,
frequency_penalty: 0,
};

export const _generateTitleConfig:ConfigInterface = {
model: defaultModel,
max_tokens: 200,//Only 6 words in title according to the prompt
temperature: 0,//for stable titles
presence_penalty: 0,
top_p: 1,
frequency_penalty: 0,
};
export const generateDefaultChat = (
title?: string,
folder?: string
Expand Down
5 changes: 1 addition & 4 deletions src/types/chat.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,10 +49,7 @@ export interface Folder {
color?: string;
}

export type ModelOptions = 'gpt-4' | 'gpt-4-32k' | 'gpt-3.5-turbo' | 'gpt-3.5-turbo-16k' ;
// | 'gpt-3.5-turbo-0301';
// | 'gpt-4-0314'
// | 'gpt-4-32k-0314'
export type ModelOptions = 'gpt-4' | 'gpt-4-32k' | 'gpt-3.5-turbo' | 'gpt-3.5-turbo-16k' | 'gpt-3.5-turbo-0301'|'gpt-3.5-turbo-0613'| 'gpt-4-0314'| 'gpt-4-32k-0314'

export type TotalTokenUsed = {
[model in ModelOptions]?: {
Expand Down