{"id":116,"date":"2023-12-28T19:01:00","date_gmt":"2023-12-28T11:01:00","guid":{"rendered":"https:\/\/blog.liu-qi.cn\/?p=116"},"modified":"2026-04-18T21:33:39","modified_gmt":"2026-04-18T13:33:39","slug":"%e8%af%b4%e4%b8%a4%e4%b8%aa%e5%9b%bd%e5%86%85%e7%bd%91%e7%bb%9c%e7%8e%af%e5%a2%83%e7%9b%b4%e6%8e%a5%e6%b3%a8%e5%86%8c%e4%bd%bf%e7%94%a8chatgpt%e7%9a%84%e6%96%b9%e6%b3%95%ef%bc%8c%e8%87%aa%e5%b7%b1","status":"publish","type":"post","link":"https:\/\/en.blog.liu-qi.cn\/2023\/12\/28\/%e8%af%b4%e4%b8%a4%e4%b8%aa%e5%9b%bd%e5%86%85%e7%bd%91%e7%bb%9c%e7%8e%af%e5%a2%83%e7%9b%b4%e6%8e%a5%e6%b3%a8%e5%86%8c%e4%bd%bf%e7%94%a8chatgpt%e7%9a%84%e6%96%b9%e6%b3%95%ef%bc%8c%e8%87%aa%e5%b7%b1\/","title":{"rendered":"Two Methods for Registering and Using ChatGPT Directly in China's Domestic Network Environment"},"content":{"rendered":"
    \n
  1. PandoraNext<\/strong><\/li>\n<\/ol>\n

    PandoraNext is probably the easiest way to casually use the web version of GPT3.5.<\/p>\n

    You can directly register an OpenAI account on PandoraNext’s mirror demo site. The entire registration and usage process has no extra requirements for your network environment; as long as you can access the internet normally, it’s fine. The account registered through it is the same as the basic account bought on Taobao.<\/p>\n

    If you only want to use the web version of GPT3.5, simply open PandoraNext’s demo site, register, and use it just like any domestic website. The only difference is that you need to copy and paste the verification email link (which you should normally click yourself) to let the proxy complete it for you.<\/p>\n

    Demo site address:<\/p>\n

    https:\/\/chat.oaifree.com\/<\/p>\n

    Starting from registration, once registered, you can use it directly.<\/p>\n

    The web version has its advantages; it’s faster than the free OpenAI API and responds more lively, not as rigid as the API.<\/p>\n

    Of course, using PandoraNext directly on the demo site’s web is the simplest and most basic usage, but it can actually do much more. You can deploy it locally on Windows, on a VPS with Linux, or via Docker on a NAS.<\/p>\n

    For example, I’ve deployed it on my ZSpace NAS and can quickly access it remotely from multiple locations using ZSpace’s built-in remote access feature.<\/p>\n

    \"\"<\/p>\n

    For other usage, you can check the project address yourself:<\/p>\n

    https:\/\/github.com\/pandora-next\/deploy?tab=readme-ov-file<\/p>\n

    And the author’s personal blog site:<\/p>\n

    https:\/\/zhile.io<\/p>\n

    Since I already had a balance topped up for the OpenAI API, I only did a simple web deployment of PandoraNext, so at least the free GPT3.5 conversations won’t consume my quota.<\/p>\n

    2.Wildcard<\/strong><\/p>\n

    Wildcard is essentially a virtual foreign currency credit card service provider, but it has integrated some features specifically for registering ChatGPT\/OpenAI API. For example, it provides a US remote network environment for registering ChatGPT (sometimes the VPN you buy might not successfully access OpenAI’s website), a UK phone number needed for topping up OpenAI, an overseas email address, and even manual binding assistance in case binding fails, etc. The website also has comprehensive tutorials on registration and top-up parts.<\/p>\n

    The Wildcard website charges a $15.99 card issuance fee and a handling fee for each top-up. I mainly use it to get an OpenAI API key and top up my OpenAI wallet balance. If you only want to use ChatGPT3.5, the zero-cost PandoraNext solution mentioned above is the most suitable.<\/p>\n

    You can use the invitation code HAPPY88 to get a 12% discount on the card issuance fee.<\/p>\n

    Or directly use this link to access the website and register for the card:<\/p>\n

    https:\/\/bewildcard.com\/i\/HAPPY88<\/p>\n

    After successfully issuing the card, follow the tutorials for the next steps. You can upgrade to a ChatGPT Plus account or obtain an OpenAI API key.<\/p>\n

    \"\"<\/p>\n

    After obtaining the OpenAI API key, the usage scenarios are much broader than just directly accessing the official website chat via a web browser. By default, a $5 balance is credited, and the API key can be used right away, but I recommend adding $10 to it for faster output speeds; with only the free quota, it will be very slow.<\/p>\n

    For example, using another chat image deployed via Docker on my NAS, with the API key, I can select the corresponding model\u2014either GPT-3.5 or GPT-4 (different models have different billing rates, of course).<\/p>\n

    \"\"<\/p>\n

    It can also be used with Stable Diffusion to have GPT generate prompts for image generation. Here, you can similarly choose whether to use GPT-3.5 or GPT-4 as the model.<\/p>\n

    \"\"<\/p>\n

    Don’t rush\u2014even if you have obtained the API key, if you are in a purely domestic network environment, you will most likely still be unable to use the OpenAI API normally.<\/p>\n

    So remember to change the official API endpoint to a third-party proxy in the API configuration, as shown in the image below:<\/p>\n

    \"\"<\/p>\n

    Replace the official endpoint domain api.openai.com in https:\/\/api.openai.com\/v1 with the third-party proxy domain api.openai-proxy.com, so the entire URL becomes https:\/\/api.openai-proxy.com\/v1. This will allow it to work within the domestic network environment.<\/p>\n

    There are many free API proxies available; besides the one mentioned above, you can also search online for a suitable proxy. Alternatively, you could set up a ‘magic’ environment at home using a software router or similar, and use the official API directly. The API endpoint is not as strictly restricted as the official website, and even Hong Kong IPs can be used normally.<\/p>\n

    Cover AI Girl:<\/p>\n

    \"\"<\/p>\n","protected":false},"excerpt":{"rendered":"

    This article details two practical methods for registering and using ChatGPT directly within China’s domestic network environment, focusing on simplicity and cost-effectiveness.<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[24],"tags":[],"class_list":["post-116","post","type-post","status-publish","format-standard","hentry","category-articles"],"_links":{"self":[{"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/posts\/116","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/comments?post=116"}],"version-history":[{"count":0,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/posts\/116\/revisions"}],"wp:attachment":[{"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/media?parent=116"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/categories?post=116"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/tags?post=116"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}