Tutorial
Tutorial posts from the AIHubMix team.
Claude Code settings.json and Environment Variables: Base URL, Auth Token and ModelClaude Code takes its configuration from JSON settings files and from environment variables. To run it through a gateway instead of Anthropic's API, you need two variables, ANTHROPIC_BASE_URL and a token, and usually a model. The variables are simple. What confuses people is where to put them, because the same variable can sit in your shell profile, in several settings.json files and in VS Code's settings. This guide starts with a setup that works, then explains where the files live, which val
MiniMax M3.1 Flash Effort Levels: Why the Default Cost 7× Low on the Same Bug FixMiniMax M3.1 Flash has five thinking levels and no off switch. If you don't pick a level, the model picks the deepest one for you. MiniMax's text generation guide says effort "defaults to max when omitted." That default is expensive in a way that's easy to miss. AIHubMix ran the same two-bug fix at every level. The default used about 10× the thinking tokens of low and took about 6× as long, and every run at every level passed all 215 tests. The model's thinking cost rose with the effort settin
Moving Off MiniMax M2.5/M2.7 Before Nov 18: Fixing M3.1 Flash's "requires adaptive thinking" 400MiniMax is retiring M2.5 and M2.7 on 18 November 2026, and the switch has already started. Since 19 October, the two older families have been closed to new users, and plan requests that name them are served by M3.1 Flash. If you're on pay-as-you-go, you have four weeks to change the model ID yourself. Most M2.7 code runs on M3.1 Flash unchanged. The trouble comes from one design choice: M3.1 Flash always thinks, and it thinks at the deepest level unless you say otherwise. That produces one lou
How to Set Up Codex config.toml: Custom API, Proxy, and Common ErrorsCodex keeps its settings in one file, config.toml: which model to use, which API to call, and how much the agent may do without asking. The format changed several times in 2026, and many guides from earlier in the year show settings that now stop Codex from starting. This guide gives you a working config first, then the parts most people change: profiles, project settings, the sandbox, and a proxy. At the end is a list of the errors you're likely to see, with what each one means. Checked again
Migrating to DeepSeek V4.1 Flash 1017: The reasoning_content 400, the Silent Alias Swap, and Peak-Hour BillingMost migrations start with changing a model name. DeepSeek V4.1 Flash 1017 did it the other way round: on October 17, 2026, DeepSeek kept the name deepseek-flash and changed the model behind it. Anyone calling deepseek-flash, deepseek-v4-flash, or deepseek-v4-flash-vision-exp on DeepSeek's API was moved over without touching a line of code. On AIHubMix the new checkpoint has its own ID, deepseek-v4.1-flash-1017, so you choose when to move. Either way, these are the problems to check: Prob
Qwen3.8 Free on AIHubMix: Fix 429s, "Rate Limited by Provider", and a Drained Token PoolMost problems with qwen3.8-max-free and qwen3.8-flash-free come from three places: the free-tier caps, the campaign capacity GMI Cloud has set aside upstream, and how quickly thinking tokens and agent context drain a 1M-token daily pool. Some of these show up as errors. Others show up as an odd-looking answer, or an empty quota by lunchtime. Alongside them, the everyday API errors still apply: a wrong key or model ID, a context that's too long, a broken tool-call history, JSON mode, content mode
x402 Tutorial: Let Your AI Agent Top Up AIHubMix with USDCAIHubMix now accepts USDC top-ups. Your AI agent can add credit to its own account with the USDC in its wallet: no web page to open, and no one has to click a button. With x402, a top-up can be as small as $1. The agent tops up as it goes, so you don't have to load a large balance in advance. It takes two steps: get a token, a wallet, and some USDC ready on the web and in an app, then run our ready-made top-up script with one command. About 20 seconds after payment, the credit shows up in yo
Moving to Gemini 4 Argon from Gemini 3.1 Pro or 3.8 Flash: What Breaks and How to Fix ItGemini 4 Argon became generally available on October 13, 2026, under the model ID gemini-4-argon, on the paid Gemini API, on Vertex AI and on AIHubMix. AIHubMix lists it at $2 per million input tokens, $10 per million output tokens and $0.10 per million cached input tokens, the same as Google's introductory rate. Google's rate runs until November 30; from December 1 it doubles to $4 and $20. [TBD: AIHubMix price from December 1] Changing the model string is the easy part. Google's What's new i
Gemini 4 Argon Thinking Levels and Long Decode: Defaults, Costs and When to Use EachGemini 4 Argon became generally available on October 13, 2026, and with it Google documented the two settings that decide what an Argon request costs and how long it takes: the thinking level and the output limit. Argon offers three thinking levels, low, medium and high, and defaults to medium. Responses stop at 64K tokens unless you turn on long decode, which lets a single response run to 1M tokens. There's a catch in the defaults. Every published Argon benchmark ran at high: Google evaluated
AIHubMix Supports JanitorAI Roleplay Chat, with Zero Data Retention on by DefaultAIHubMix supports roleplay chat. Connect it to JanitorAI as a proxy, pick a free model, and your characters reply through AIHubMix. Every AIHubMix account also has Zero Data Retention (ZDR) turned on by default, so we don't log, store, or keep your messages, your character prompts, or the model's replies. The short version: * Roleplay chat: AIHubMix connects to JanitorAI as a proxy. Dozens of free models are available, and our first pick for roleplay chat is xiaomi-mimo-v2.6-pro-free. * Ze