🚀 Socket Launch Week 🚀 Day 5: Introducing Socket Fix.Learn More
Socket
Sign inDemoInstall
Socket

ai-fallback

Package Overview
Dependencies
Maintainers
0
Versions
9
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

ai-fallback

Automatically switch AI SDK model provider when one of the providers has downtime

0.0.3
Source
npm
Version published
Weekly downloads
1.4K
24.46%
Maintainers
0
Weekly downloads
 
Created
Source

AI Fallback

Automatically switch between AI model providers when one experiences downtime or errors.

Why?

AI providers can encounter:

  • Rate limiting
  • Service outages
  • API errors
  • Capacity issues
  • Timeouts

This package ensures reliability by specifying multiple AI models as fallbacks. It automatically switches to the next available model if the primary fails, maintaining application uptime.

Installation

npm install ai-fallback

Models Fallback Reset

Reset to the primary model after a delay (e.g., 1 minute):

const model = createFallback({
    models: [anthropic('claude-3-haiku-20240307'), openai('gpt-3.5-turbo')],
    resetAfterMs: 60000, // Reset after 1 minute
    onError: (error, modelId) => {
        console.error(`Error with model ${modelId}:`, error)
    },
    modelResetInterval: 60000, // Reset to first model after 1 minute of the first error
})

Usage

Create a Fallback Model

import { createFallback } from 'ai-fallback'
import { createOpenAI } from '@ai-sdk/openai'
import { createAnthropic } from '@ai-sdk/anthropic'

const model = createFallback({
    models: [
        createAnthropic('claude-3-haiku-20240307'),
        createOpenAI('gpt-3.5-turbo'),
    ],
})

Text Generation

Generate text with automatic fallback:

const result = await generateText({
    model,
    system: 'You are a helpful assistant.',
    messages: [{ role: 'user', content: 'Count from 1 to 5.' }],
})

Streaming Responses

Stream text responses:

const stream = await streamText({
    model,
    system: 'You are a helpful assistant.',
    messages: [{ role: 'user', content: 'Count from 1 to 5.' }],
})

for await (const chunk of stream.textStream) {
    console.log(chunk)
}

Structured Output

Stream typed objects using Zod schemas:

import { z } from 'zod'

const stream = await streamObject({
    model,
    system: 'You are a helpful assistant.',
    messages: [
        {
            role: 'user',
            content: 'Give me a person object with name and age properties.',
        },
    ],
    schema: z.object({
        name: z.string(),
        age: z.number(),
    }),
})

for await (const chunk of stream.partialObjectStream) {
    console.log(chunk)
}

FAQs

Package last updated on 28 Nov 2024

Did you know?

Socket

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Install

Related posts