mirror of
https://github.com/ItzCrazyKns/Perplexica.git
synced 2025-07-18 06:28:30 +00:00
f473a581ce
configurable delay feature. 1. Created AlternatingMessageValidator (renamed from MessageProcessor): -Focused on handling alternating message patterns -Made it model-agnostic with configuration-driven approach -Kept the core validation logic intact 2. Created ReasoningChatModel (renamed from DeepSeekChat): -Made it generic for any model with reasoning/thinking capabilities -Added configurable streaming delay parameter (streamDelay) -Implemented delay logic in the streaming process 3. Updated the DeepSeek provider: -Now uses ReasoningChatModel for deepseek-reasoner with a 50ms delay -Uses standard ChatOpenAI for deepseek-chat -Added a clear distinction between models that need reasoning capabilities Updated references in metaSearchAgent.ts: 4. Changed import from messageProcessor to alternatingMessageValidator -Updated function calls to use the new validator -The configurable delay implementation allows to control the speed of token generation, which can help with the issue you were seeing. The delay is set to 20ms by default for the deepseek-reasoner model, but you can adjust his value in the deepseek.ts provider file to find the optimal speed. This refactoring maintains all the existing functionality while making the code more maintainable and future-proof. The separation of concerns between message validation and model implementation will make it easier to add support for other models with similar requirements in the future.
91 lines
2.6 KiB
TypeScript
91 lines
2.6 KiB
TypeScript
import { ReasoningChatModel } from '../reasoningChatModel';
|
|
import { ChatOpenAI } from '@langchain/openai';
|
|
import logger from '../../utils/logger';
|
|
import { getDeepseekApiKey } from '../../config';
|
|
import axios from 'axios';
|
|
|
|
interface DeepSeekModel {
|
|
id: string;
|
|
object: string;
|
|
owned_by: string;
|
|
}
|
|
|
|
interface ModelListResponse {
|
|
object: 'list';
|
|
data: DeepSeekModel[];
|
|
}
|
|
|
|
interface ChatModelConfig {
|
|
displayName: string;
|
|
model: ReasoningChatModel | ChatOpenAI;
|
|
}
|
|
|
|
// Define which models require reasoning capabilities
|
|
const REASONING_MODELS = ['deepseek-reasoner'];
|
|
|
|
const MODEL_DISPLAY_NAMES: Record<string, string> = {
|
|
'deepseek-reasoner': 'DeepSeek R1',
|
|
'deepseek-chat': 'DeepSeek V3'
|
|
};
|
|
|
|
export const loadDeepSeekChatModels = async (): Promise<Record<string, ChatModelConfig>> => {
|
|
const deepSeekEndpoint = 'https://api.deepseek.com';
|
|
|
|
const apiKey = getDeepseekApiKey();
|
|
if (!apiKey) return {};
|
|
|
|
if (!deepSeekEndpoint || !apiKey) {
|
|
logger.debug('DeepSeek endpoint or API key not configured, skipping');
|
|
return {};
|
|
}
|
|
|
|
try {
|
|
const response = await axios.get<{ data: DeepSeekModel[] }>(`${deepSeekEndpoint}/models`, {
|
|
headers: {
|
|
'Content-Type': 'application/json',
|
|
'Authorization': `Bearer ${apiKey}`,
|
|
},
|
|
});
|
|
|
|
const deepSeekModels = response.data.data;
|
|
|
|
const chatModels = deepSeekModels.reduce<Record<string, ChatModelConfig>>((acc, model) => {
|
|
// Only include models we have display names for
|
|
if (model.id in MODEL_DISPLAY_NAMES) {
|
|
// Use ReasoningChatModel for models that need reasoning capabilities
|
|
if (REASONING_MODELS.includes(model.id)) {
|
|
acc[model.id] = {
|
|
displayName: MODEL_DISPLAY_NAMES[model.id],
|
|
model: new ReasoningChatModel({
|
|
apiKey,
|
|
baseURL: deepSeekEndpoint,
|
|
modelName: model.id,
|
|
temperature: 0.7,
|
|
streamDelay: 50, // Add a small delay to control streaming speed
|
|
}),
|
|
};
|
|
} else {
|
|
// Use standard ChatOpenAI for other models
|
|
acc[model.id] = {
|
|
displayName: MODEL_DISPLAY_NAMES[model.id],
|
|
model: new ChatOpenAI({
|
|
openAIApiKey: apiKey,
|
|
configuration: {
|
|
baseURL: deepSeekEndpoint,
|
|
},
|
|
modelName: model.id,
|
|
temperature: 0.7,
|
|
}),
|
|
};
|
|
}
|
|
}
|
|
return acc;
|
|
}, {});
|
|
|
|
return chatModels;
|
|
} catch (err) {
|
|
logger.error(`Error loading DeepSeek models: ${String(err)}`);
|
|
return {};
|
|
}
|
|
};
|