How to Connect NextChat (ChatGPT-Next-Web) to APIVAI

Connect NextChat to APIVAI

NextChat (ChatGPT-Next-Web) is one of the most popular
one-click-deploy ChatGPT-style web UIs. It speaks the OpenAI-compatible format, so you can run it on
APIVAI's Claude and GPT models by pointing the API endpoint at APIVAI — in the UI or via env vars.

Option A — in Settings

1. Open NextChat → Settings.
2. Find Custom Endpoint (自定义接口) and enable it.
3. Set:
   • API Endpoint / 接口地址: https://api.apivai.com
   • API Key / 密钥: your APIVAI key
4. (Optional) set Custom Models to the names you want.
5. Save, start a chat, and pick a model.

Option B — environment variables (self-hosted / Vercel / Docker)

OPENAI_API_KEY=your_apivai_key
BASE_URL=https://api.apivai.com
# optional: expose specific models
CUSTOM_MODELS=+claude-sonnet-4-6,+gpt-5.5

Redeploy/restart; NextChat now routes through APIVAI.

Confirm models

curl -s https://api.apivai.com/v1/models \
  -H "Authorization: Bearer $APIVAI_API_KEY"

Use a returned name in Custom Models.

Troubleshooting

• 401 / no response — check the endpoint (https://api.apivai.com) and key.
• model_not_found — set a name from /v1/models in Custom Models.
• Wrong base path — NextChat appends /v1/...; use the host, not /v1, in BASE_URL.

Why APIVAI for NextChat

One endpoint gives NextChat both Claude and GPT models at a fraction of list price, pay-as-you-go,
with crypto/USDT/Alipay and no VPN — ideal for a self-hosted personal or team chat.

FAQ

Does NextChat work with APIVAI? Yes — set the custom API endpoint to https://api.apivai.com
and your APIVAI key, in Settings or via BASE_URL/OPENAI_API_KEY.

Can I use Claude and GPT both? Yes — one key, both families; list them in Custom Models.

Do I need to change code? No — it's a settings/env change.

Get started

Point NextChat's endpoint at APIVAI, add your key, set a model from /v1/models. Examples:
APIVAI examples repo.