Ian Macartney ianmacartney

ianmacartney / llm.ts

Last active December 5, 2024 16:30

Implementation of chat completions and embeddings for any OpenAI-compliant services, using browser fetch and no imports/dependencies

	export const CONFIG = {
	// Together AI:
	url: "https://api.together.xyz",
	chatModel: "meta-llama/Llama-3-8b-chat-hf",
	embeddingModel: "togethercomputer/m2-bert-80M-8k-retrieval", // dim 768
	// OpenAI:
	// url: "https://api.openai.com",
	// chatModel: "gpt-4o",
	// embeddingModel: "text-embedding-ada-002", // dim 1536
	};