import type { AxModelInfo } from '../types.js'; import { AxAIWebLLMModel } from './types.js'; /** * WebLLM model information * Note: WebLLM runs models locally in the browser, so there are no API costs * However, we include context window and capability information */ export const axModelInfoWebLLM: AxModelInfo[] = [ // Llama 3.1 series { name: AxAIWebLLMModel.Llama31_8B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, // Local inference - no cost completionTokenCostPer1M: 0, // Local inference - no cost contextWindow: 128000, maxTokens: 4096, }, { name: AxAIWebLLMModel.Llama31_70B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 128000, maxTokens: 4096, isExpensive: true, // Large model - requires significant compute }, // Llama 3.2 series { name: AxAIWebLLMModel.Llama32_1B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 128000, maxTokens: 2048, }, { name: AxAIWebLLMModel.Llama32_3B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 128000, maxTokens: 2048, }, // Mistral series { name: AxAIWebLLMModel.Mistral7B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 32768, maxTokens: 4096, }, // Phi series { name: AxAIWebLLMModel.Phi35_Mini_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 128000, maxTokens: 4096, }, // Gemma series { name: AxAIWebLLMModel.Gemma2_2B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 8192, maxTokens: 2048, }, { name: AxAIWebLLMModel.Gemma2_9B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 8192, maxTokens: 2048, }, // Qwen series { name: AxAIWebLLMModel.Qwen2_5_0_5B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 32768, maxTokens: 2048, }, { name: AxAIWebLLMModel.Qwen2_5_1_5B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 32768, maxTokens: 2048, }, { name: AxAIWebLLMModel.Qwen2_5_3B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 32768, maxTokens: 2048, }, { name: AxAIWebLLMModel.Qwen2_5_7B_Instruct, currency: 'usd', promptTokenCostPer1M: 0, completionTokenCostPer1M: 0, contextWindow: 32768, maxTokens: 4096, }, ];