API application cases in various business scenarios
How can you make LiteLLM orchestrate multiple Large Language Models like OpenAI, Claude, Gemini, and DeepSeek simultaneously without getting blocked by overseas account, network, or payment issues? The answer is to connect LiteLLM to an OpenAI-compatible API proxy service. In this article, we’ll use LiteLLM + APIYI (apiyi.com) as an example to walk you through…
title: "RikkaHub Configuration Guide: Connecting to APIYI and Mastering API Channel Types" description: "Learn how to correctly configure RikkaHub with APIYI. We break down the differences between /v1 and /v1beta channels to help you access all LLMs seamlessly." tags: [RikkaHub, APIYI, LLM, Android, Tutorial] Author's Note: RikkaHub is an Android client that supports multiple LLMs….
Let’s start with the bottom line: The first-generation Nano Banana (gemini-2.5-flash-image) definitely has some hard limitations—the resolution is capped at 1K, and text often comes out garbled. That’s just a fact. In an era where Nano Banana Pro can output high-quality 2K images and Nano Banana 2 can handle 4K, the first generation has clearly…
If you've encountered this error while connecting Claude Code to AWS Bedrock: InvokeModelWithResponseStream: operation error Bedrock Runtime: InvokeModelWithResponseStream, https response error StatusCode: 400, RequestID: 47574f13-c884-4b12-a003-6d0cf252d8dd, ValidationException: system.1.cache_control.***.scope: Extra inputs are not permitted This happens frequently, especially when using –resume to pick up a previous session—this is a known compatibility issue, not a mistake in your…
Ever run into this headache? You're building a project that uses OpenAI's GPT, Anthropic's Claude, and Google's Gemini, but every model has a different SDK, a different API format, and even different error-handling logic. One model swap means rewriting half your codebase. That’s exactly what LiteLLM solves. Simply put, LiteLLM is the "universal translator" for…
When using OpenClaw, many users run into a common question: "The current model is too expensive or just not the right fit—how do I switch to another one?" For example, if you're currently using the claude-sonnet-4-6 via Anthropic Messages mode and burning through tokens, you might want to switch to the more budget-friendly gpt-5.4-mini. Do…
description: Discover how to use Positron IDE for data science, integrate Claude Code, and save 20% on Claude API costs using APIYI. Author's Note: This guide introduces the positioning and core features of Positron IDE, explains how to use Claude Code within the Positron terminal, and shows how to configure APIYI environment variables to get…
Author's Note: A complete guide to the Claude Cowork experience for newcomers, detailing what it’s actually like to have an AI control your desktop, what it can and can't do, and the core differences between Cowork (desktop GUI) and Claude Code (command-line CLI). "AI, help me operate my computer"—it sounds like science fiction, but Claude…
Author's Note: Without switching to cheaper, lower-quality model channels, this guide details how OpenClaw controls input token length to save costs. We'll cover 6 key strategies: task isolation for new conversations, precise code block retrieval instead of full-file ingestion, context pruning, local QMD search, and more. OpenClaw is notorious for being a "token burner"—some users…
Author's Note: A detailed explanation of the complete workflow for generating flowcharts from text using AI: Step 1 uses a Large Language Model to organize structured content, Step 2 uses image models like Nano Banana Pro to generate professional flowcharts. "I have a bunch of text descriptions and want to generate a nice-looking flowchart. Which…
