DEV Community: APIVAI

How to Connect Roo Code to APIVAI (Cheap API for an AI Coding Agent)

APIVAI — Sun, 28 Jun 2026 20:07:22 +0000

Connect Roo Code to APIVAI

Roo Code (formerly Roo Cline) is a popular autonomous
coding agent for VS Code. It supports OpenAI-compatible providers, so you can run it on APIVAI's
Claude and GPT models — agentic edits cost a lot of tokens, so a cheap compatible gateway matters.

Configure the provider

Open Roo Code's settings (the gear icon in the Roo panel).
API Provider: choose OpenAI Compatible.
Set:
- Base URL: https://api.apivai.com/v1
- API Key: your APIVAI key
- Model: a name APIVAI serves (e.g. claude-sonnet-4-6 or gpt-5.5)
Save.

Roo Code now routes its requests through APIVAI.

Pick a model

Confirm valid names:

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

Claude Sonnet is a strong default for agentic coding; use it for multi-step edits and reasoning.

Tips for agent workloads

Agents send many large requests — a discounted gateway like APIVAI noticeably lowers the bill.
Prefer a model with a large context window for big files/repos.
Make sure streaming passes through so you see progress as the agent works.

Troubleshooting

401 — check the key and that the provider is "OpenAI Compatible" with Base URL ending /v1.
model_not_found — use a name from /v1/models.
Truncated output — raise max tokens / pick a larger-context model.

FAQ

Does Roo Code work with APIVAI? Yes — select "OpenAI Compatible", set Base URL
https://api.apivai.com/v1 and your APIVAI key.

Which model for Roo Code? Claude Sonnet for agentic coding and reasoning; GPT-5.5 for faster,
cheaper runs.

Why use a gateway for an agent? Coding agents burn tokens; APIVAI's discounted, OpenAI-compatible
access cuts the cost without changing your workflow.

Get started

Set the OpenAI-compatible provider in Roo Code with APIVAI's base URL + key and a model from
/v1/models. Examples: APIVAI examples repo.

How to Build a Telegram AI Bot with APIVAI (Cheap Claude & GPT)

APIVAI — Sun, 28 Jun 2026 20:07:18 +0000

Build a Telegram AI bot with APIVAI

A Telegram AI bot is one of the quickest ways to ship an AI assistant: create a bot with BotFather,
run a small script that forwards messages to an OpenAI-compatible model, and reply. APIVAI provides
the model (Claude or GPT) at a fraction of list price, with crypto/USDT/Alipay payment.

This guide is the full working bot.

1. Create the bot

Message @botfather on Telegram → /newbot → choose a name → copy the bot token.

2. The bot (Node.js, long polling — no server needed)

npm install node-telegram-bot-api openai

import TelegramBot from "node-telegram-bot-api";
import OpenAI from "openai";

const bot = new TelegramBot(process.env.TELEGRAM_BOT_TOKEN, { polling: true });
const ai = new OpenAI({ apiKey: process.env.APIVAI_API_KEY, baseURL: "https://api.apivai.com/v1" });
const SYSTEM = "You are a helpful assistant. Answer concisely in the user's language.";

const history = new Map(); // chatId -> messages[]

bot.on("message", async (msg) => {
  if (!msg.text || msg.text.startsWith("/")) return;
  const chatId = msg.chat.id;
  const msgs = history.get(chatId) || [{ role: "system", content: SYSTEM }];
  msgs.push({ role: "user", content: msg.text });

  bot.sendChatAction(chatId, "typing");
  const r = await ai.chat.completions.create({ model: "gpt-5.5", messages: msgs.slice(-12), max_tokens: 500 });
  const reply = r.choices[0].message.content;

  msgs.push({ role: "assistant", content: reply });
  history.set(chatId, msgs.slice(-12)); // keep recent context
  bot.sendMessage(chatId, reply);
});

Set TELEGRAM_BOT_TOKEN and APIVAI_API_KEY, run node bot.js, and message your bot.

3. Add commands & polish

/start → a welcome message; /reset → clear that chat's history.
Trim history (last ~12 messages) to control token cost.
Stream long answers by editing the message as chunks arrive (optional).
For groups, only respond when mentioned or replied to.

Pick the model

GPT-5.5 for fast, natural, multilingual replies; Claude Sonnet for coding/long-reasoning bots.
Switch by changing the model string — it's OpenAI-compatible.

FAQ

Can a Telegram bot use APIVAI? Yes — it's a normal OpenAI chat call with the base URL set to
https://api.apivai.com/v1; the bot forwards messages and returns the reply.

Do I need a server? No — long polling (above) runs anywhere Node runs. Use webhooks if you want
a serverless deployment.

Which model should the bot use? GPT-5.5 for general/multilingual chat; Claude Sonnet for
code-heavy bots.

How do I control cost? Trim conversation history and cap max_tokens; APIVAI's per-token price
is already low.

Get started

Create a bot with BotFather, drop in the script with your APIVAI key, and run it. Examples:
APIVAI examples repo.

The Cheapest Way to Use Claude & GPT APIs in 2026

APIVAI — Thu, 25 Jun 2026 15:46:00 +0000

The cheapest way to use Claude & GPT in 2026

If you run AI coding agents or apps all day, the API bill is usually your biggest cost. The good news: you rarely need to pay official list price. Because almost every tool speaks the OpenAI-compatible API format, the model provider is just a configuration value — you can route to a cheaper compatible gateway without changing any code.

This guide covers the cheapest, most practical way to access Claude and GPT models in 2026.

Why it works: OpenAI-compatible everywhere

Claude Code, Cursor, Cline, Codex CLI, and the official OpenAI/Anthropic SDKs all talk to an OpenAI-compatible endpoint. That means switching providers is two environment variables:

export OPENAI_BASE_URL="https://api.apivai.com/v1"
export OPENAI_API_KEY="sk-your-key"

The tool doesn't care who is behind the URL, as long as it speaks the standard format.

What to look for in a cheap provider

Factor	Why it matters
Price vs official	The whole point — aim for a large discount on input/output tokens
OpenAI-compatible	Drop-in for existing tools, no rewrite
Streaming passthrough	Token-by-token output must keep working (real SSE, not buffered)
Model coverage	Both Claude and GPT families behind one key
Pay-as-you-go	No subscription; top up and spend down
Payment options	Cards, plus crypto/USDT/Alipay if you can't use a card

How APIVAI fits

APIVAI is an OpenAI- and Anthropic-compatible gateway to Claude and GPT models at a fraction of list price. One key works across models, streaming is proxied as real chunks, and you can pay with crypto, USDT, or Alipay — no VPN, no subscription.

curl https://api.apivai.com/v1/chat/completions \
  -H "Authorization: Bearer $APIVAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"claude-sonnet-4-6","messages":[{"role":"user","content":"Hello"}]}'

Cut costs further

Pick the right model per task — use a smaller/faster model for routine work and reserve the most powerful model for hard problems.
Cache where you can; reuse context instead of resending it.
Watch per-model context limits so you aren't paying for tokens you don't need.

Get started

Create a key, call GET /v1/models to see what's available, set the two environment variables, and send your first request. Your existing tools keep working — just cheaper.

How to Use Continue.dev with a Cheap OpenAI-Compatible API

APIVAI — Thu, 25 Jun 2026 15:45:56 +0000

Use Continue.dev with a cheap OpenAI-compatible API

Continue.dev is a popular open-source AI assistant for VS Code and JetBrains. Like most modern AI coding tools, it speaks the OpenAI-compatible API format — so you don't have to pay official list price. Point it at a cheaper compatible provider and everything works unchanged.

This guide shows the exact Continue config to use an OpenAI-compatible gateway like APIVAI.

Why it works

Continue lets you define any model with a provider: "openai" block and a custom apiBase. That's all it takes to route through a discount gateway — the requests are standard OpenAI chat completions.

Configure Continue

Open your Continue config (~/.continue/config.json, or the YAML config in newer versions) and add a model:

{
  "models": [
    {
      "title": "APIVAI - Claude Sonnet",
      "provider": "openai",
      "model": "claude-sonnet-4-6",
      "apiBase": "https://api.apivai.com/v1",
      "apiKey": "sk-your-apivai-key"
    }
  ]
}

That's it. Reload Continue and the model appears in the dropdown. Send a message to confirm it streams.

Pick a model that exists

Model availability varies, so list what's available first instead of guessing:

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

Use a model name from that response for the model field.

Things to verify

Streaming — Continue streams responses; make sure your provider passes real SSE chunks so output appears token by token.
Model name — must match one the provider serves (avoids model_not_found).
Context window — pick a model whose context fits your codebase chunks.

Why APIVAI

APIVAI is OpenAI- and Anthropic-compatible, covers Claude and GPT models behind one key, and is priced at a fraction of official list. Pay-as-you-go with crypto, USDT, or Alipay, no VPN required — handy if you can't or don't want to use a card.

Get started

Grab a key, drop the config block above into Continue with a model from /v1/models, and start coding at a fraction of the cost — no code changes, same workflow.

How to Run Whisper Locally for Real-Time Speech-to-Text (faster-whisper)

APIVAI — Thu, 25 Jun 2026 15:45:29 +0000

Run Whisper locally for real-time speech-to-text

For live translation and voice apps, you want speech-to-text that's fast and free per use — which
means running Whisper locally with faster-whisper. It pairs perfectly with APIVAI for the
translation step: Whisper turns audio into text locally, APIVAI (GPT-5.5) translates it cheaply over
an OpenAI-compatible call, and you keep latency low and per-second cost at zero for the audio part.

This guide gets local Whisper running for real-time use.

Install

pip install faster-whisper

faster-whisper is a fast reimplementation of OpenAI's Whisper using CTranslate2. It runs on GPU
(CUDA) or CPU.

Choose a model size

Model	Speed	Accuracy	Use
`tiny` / `base`	fastest	lower	quick captions, weak hardware
`small`	fast	good	recommended for live
`medium`	slower	better	accuracy over speed
`large-v3`	slowest	best	offline/high-accuracy jobs

For live captioning, small on a GPU is the sweet spot.

Transcribe in near real time

from faster_whisper import WhisperModel

model = WhisperModel("small", device="cuda", compute_type="float16")  # device="cpu", compute_type="int8" for CPU

def transcribe(audio_path: str, lang: str = "zh") -> str:
    segments, _ = model.transcribe(audio_path, language=lang, vad_filter=True, beam_size=1)
    return "".join(s.text for s in segments).strip()

Capture short audio chunks (1–3s) from the mic/OBS monitor and feed them in.
vad_filter=True skips silence so you don't transcribe noise.
beam_size=1 favors speed for live use.

Hook it to translation (APIVAI)

from openai import OpenAI
ai = OpenAI(api_key="YOUR_APIVAI_API_KEY", base_url="https://api.apivai.com/v1")

def translate(text, target="Spanish"):
    r = ai.chat.completions.create(model="gpt-5.5", messages=[
        {"role": "system", "content": f"Translate to natural {target}. Output only the translation."},
        {"role": "user", "content": text}])
    return r.choices[0].message.content.strip()

print(translate(transcribe("chunk.wav")))

Performance tips

GPU (float16) is much faster than CPU; on CPU use int8 and a smaller model.
Pre-load the model once and reuse it; don't reload per chunk.
Keep chunks short; longer audio adds latency.

FAQ

Is local Whisper free? Yes — faster-whisper runs on your own hardware; there's no per-call
cost. You only pay for the translation step (e.g. GPT-5.5 via APIVAI), which is cheap.

Does APIVAI provide Whisper? No — APIVAI provides the OpenAI/Anthropic-compatible chat models
(the translation step). Run Whisper locally for speech-to-text.

Which model size for live captions? small on a GPU — a good balance of speed and accuracy.

What latency can I expect? With small + short chunks, transcription is sub-second; end-to-end
(STT → translate → caption) is typically ~1–2s.

Get started

Install faster-whisper, run the transcribe loop, and send the text to APIVAI for translation.
Examples: APIVAI examples repo.

How to Build a Discord AI Bot with APIVAI (Cheap Claude & GPT)

APIVAI — Thu, 25 Jun 2026 15:45:26 +0000

Build a Discord AI bot with APIVAI

A Discord AI bot lets your server chat with Claude or GPT. Create a bot application, run a small
discord.js script that forwards messages to an OpenAI-compatible model, and reply. APIVAI provides
the model at a fraction of list price, with crypto/USDT/Alipay payment.

1. Create the bot

Go to the Discord Developer Portal → New Application → Bot → copy the bot token.
Enable the Message Content Intent (Bot settings) so the bot can read messages.
Invite the bot to your server (OAuth2 URL with the bot scope and Send Messages permission).

2. The bot (discord.js)

npm install discord.js openai

import { Client, GatewayIntentBits } from "discord.js";
import OpenAI from "openai";

const client = new Client({ intents: [GatewayIntentBits.Guilds, GatewayIntentBits.GuildMessages, GatewayIntentBits.MessageContent] });
const ai = new OpenAI({ apiKey: process.env.APIVAI_API_KEY, baseURL: "https://api.apivai.com/v1" });
const SYSTEM = "You are a helpful assistant in a Discord server. Be concise.";

client.on("messageCreate", async (msg) => {
  if (msg.author.bot) return;
  // respond when mentioned (so the bot isn't noisy)
  if (!msg.mentions.has(client.user)) return;

  const prompt = msg.content.replace(/<@!?\d+>/g, "").trim();
  await msg.channel.sendTyping();
  const r = await ai.chat.completions.create({
    model: "gpt-5.5",
    messages: [{ role: "system", content: SYSTEM }, { role: "user", content: prompt }],
    max_tokens: 500,
  });
  msg.reply(r.choices[0].message.content.slice(0, 1900)); // Discord 2000-char limit
});

client.login(process.env.DISCORD_BOT_TOKEN);

Set DISCORD_BOT_TOKEN and APIVAI_API_KEY, run node bot.js, and mention the bot in your server.

Polish

Add slash commands (/ask) for a cleaner UX.
Keep short per-channel history for context; trim to control token cost.
Split replies over 2000 chars into multiple messages.
Pick GPT-5.5 for general chat, Claude Sonnet for coding help.

FAQ

Can a Discord bot use APIVAI? Yes — it's a standard OpenAI chat call with base URL
https://api.apivai.com/v1; the bot forwards the message and posts the reply.

Which model should the bot use? GPT-5.5 for general/multilingual chat; Claude Sonnet for code.

How do I keep it from being spammy? Only respond when mentioned or via a slash command.

How do I control cost? Trim history and cap max_tokens; APIVAI's per-token price is already low.

Get started

Create a bot app, drop in the script with your APIVAI key, and run it. Examples:
APIVAI examples repo.

Build an Enterprise Knowledge Base with Open WebUI + APIVAI (RAG)

APIVAI — Thu, 25 Jun 2026 15:41:22 +0000

Build an enterprise knowledge base with Open WebUI + APIVAI

You can stand up a private, self-hosted "chat with our documents" knowledge base without writing a
RAG pipeline from scratch. Open WebUI has built-in document upload and
retrieval; APIVAI supplies the language model (Claude or GPT) over one OpenAI-compatible
connection, at a fraction of official price. Your documents stay on your own server; only the
model calls go out.

This guide covers the setup and the design choices that matter.

Why this stack

Self-hosted UI + storage — documents and chat history live on infrastructure you control.
No custom RAG code — Open WebUI handles upload, chunking, embedding, and retrieval.
One model connection — APIVAI gives you Claude and GPT models behind a single key.
Cost control — pay-as-you-go at a fraction of list; crypto/USDT/Alipay accepted.

1. Run Open WebUI and connect APIVAI

Start Open WebUI (Docker), then add APIVAI as an OpenAI connection (Admin Panel → Settings →
Connections → OpenAI API):

API Base URL: https://api.apivai.com/v1
API Key: your APIVAI key

Confirm models load in the chat model selector. (See the dedicated "Connect Open WebUI to APIVAI"
guide for screenshots and troubleshooting.)

2. Add your knowledge base

In Open WebUI:

Go to Workspace → Knowledge (or Documents).
Create a collection and upload your files (PDF, Markdown, DOCX, TXT).
Open WebUI chunks and embeds them automatically.
In a chat, reference the collection with # (e.g. #company-handbook) so the model answers from those documents.

3. Choose the answering model

Claude Sonnet is a strong default for knowledge-base Q&A: large context, faithful summarization, careful answers over long source material.
GPT-5.5 is great when answers should be fast and conversational, or multilingual.

Set the model per chat or as the workspace default. Because it's OpenAI-compatible, switching is
just picking a different model name.

4. Make answers trustworthy

Cite sources: instruct the model (system prompt) to quote or reference the document section it used, and to say "not in the documents" when it isn't.
Scope retrieval: keep collections focused (one per domain: HR, product, support) so retrieval stays relevant.
Access control: use Open WebUI's user/group permissions so teams only see their collections.
Keep it fresh: re-upload documents when they change; remove stale ones.

5. Embeddings note

Open WebUI can compute embeddings locally (default) or via an external embeddings endpoint. APIVAI
focuses on the chat/completions side (Claude and GPT), so keep Open WebUI's built-in/local
embeddings for retrieval and use APIVAI for the answering model — a clean, low-cost split.

Example use cases

Internal HR/IT helpdesk over policy docs.
Product/support team answering from manuals and past tickets.
Sales enablement over spec sheets and pricing.

FAQ

Where do my documents live? On your Open WebUI server — they are not sent to APIVAI; only the
chat prompt (with retrieved snippets) is.

Which model for a knowledge base? Claude Sonnet for faithful long-context answers; GPT-5.5 for
fast multilingual replies.

Do I need to build embeddings/RAG myself? No — Open WebUI handles it; APIVAI is just the
answering model.

Get started

Run Open WebUI, connect APIVAI (https://api.apivai.com/v1 + key), upload your docs, and pick a
model. Examples: APIVAI examples repo.

How to Connect Open WebUI to APIVAI (OpenAI-Compatible)

APIVAI — Thu, 25 Jun 2026 15:41:19 +0000

Connect Open WebUI to APIVAI

Open WebUI is a popular self-hosted ChatGPT-style interface. It talks to
any OpenAI-compatible endpoint, so you can run it on top of APIVAI's Claude and GPT models by
adding one connection — base URL plus key. No plugins, no code.

This guide shows the exact steps.

1. Run Open WebUI

If you don't have it yet, the quickest start is Docker:

docker run -d -p 3000:8080 \
  -v open-webui:/app/backend/data \
  --name open-webui \
  ghcr.io/open-webui/open-webui:main

Open http://localhost:3000 and create the admin account.

2. Add APIVAI as an OpenAI connection

In Open WebUI:

Click your avatar → Admin Panel → Settings → Connections.
Under OpenAI API, click + to add a connection.
Set:
- API Base URL: https://api.apivai.com/v1
- API Key: your APIVAI key
Save, then refresh.

Open WebUI will pull the model list from APIVAI. Claude and GPT models now appear in the model
selector at the top of a new chat.

3. Pick a model that exists

If the dropdown is empty or a model errors, confirm the available names:

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

Use one of the returned names. You can also hide models you don't want from the admin model
settings.

4. Verify

Start a new chat, pick a Claude or GPT model, and send a message. You should see a streamed
response. If it streams token by token, the OpenAI-compatible connection is working end to end.

Troubleshooting

No models in the dropdown — re-check the Base URL ends with /v1 and the key is valid; refresh the page after saving.
model_not_found — call GET /v1/models and use an exact returned name.
No streaming — make sure you didn't disable streaming in the chat settings.
401 — the key is missing/incorrect or has a leading/trailing space.

Why APIVAI for Open WebUI

One OpenAI-compatible connection gives Open WebUI access to both Claude and GPT models.
A fraction of official list price, pay-as-you-go — good for a team sharing one self-hosted UI.
Crypto / USDT / Alipay payment, no VPN, no subscription.

For a private document Q&A setup, see how to build an enterprise knowledge base with Open WebUI +
APIVAI. Working API examples are in the
APIVAI examples repo.

How to Connect Chatbox to APIVAI (Cheap Claude & GPT)

APIVAI — Thu, 25 Jun 2026 15:40:50 +0000

Connect Chatbox to APIVAI

Chatbox is a popular desktop and mobile AI client. It supports
OpenAI-compatible providers, so you can use APIVAI's Claude and GPT models by setting the API host
and key — no code.

Configure

Open Chatbox → Settings.
Model Provider: choose OpenAI API (or "Add Custom Provider" → OpenAI-compatible).
Set:
- API Host / API Domain: https://api.apivai.com/v1
- API Key: your APIVAI key
Model: pick or enter a name APIVAI serves (e.g. claude-sonnet-4-6, gpt-5.5).
Save and start chatting.

Confirm available models

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

Enter one of those names if a model errors.

Troubleshooting

401 — check the API key and that the host is https://api.apivai.com/v1.
model_not_found — use an exact name from /v1/models.
No streaming — confirm streaming is on in Chatbox settings.

Why APIVAI for Chatbox

A single OpenAI-compatible host gives Chatbox both Claude and GPT models at a fraction of list
price, pay-as-you-go, with crypto/USDT/Alipay and no VPN — great for a personal cross-device client.

FAQ

Does Chatbox work with APIVAI? Yes — set the API host to https://api.apivai.com/v1 and your
APIVAI key in the OpenAI provider settings.

Can I use Claude and GPT? Yes — one key exposes both families.

Do I need any code? No — it's a settings change in the Chatbox app.

Get started

Set APIVAI's host + key in Chatbox and pick a model from /v1/models. Examples:
APIVAI examples repo.

How to Build a Slack AI Bot with APIVAI (Cheap Claude & GPT)

APIVAI — Thu, 25 Jun 2026 15:40:47 +0000

Build a Slack AI bot with APIVAI

A Slack AI bot lets your workspace ask Claude or GPT from any channel. Create a Slack app, subscribe
to message events, and forward them to an OpenAI-compatible model. APIVAI provides the model cheaply.

1. Create the Slack app

api.slack.com/apps → Create New App → enable Bot Token Scopes (app_mentions:read, chat:write) → install to your workspace → copy the Bot Token (xoxb-...).
Enable Event Subscriptions and subscribe to app_mention.

2. The server (Node)

import express from "express";
import OpenAI from "openai";
const app = express(); app.use(express.json());
const ai = new OpenAI({ apiKey: process.env.APIVAI_API_KEY, baseURL: "https://api.apivai.com/v1" });

app.post("/slack/events", async (req, res) => {
  if (req.body.type === "url_verification") return res.send(req.body.challenge); // Slack handshake
  res.sendStatus(200);
  const e = req.body.event;
  if (e?.type === "app_mention") {
    const prompt = e.text.replace(/<@[^>]+>/g, "").trim();
    const r = await ai.chat.completions.create({ model: "gpt-5.5", messages: [{ role: "user", content: prompt }], max_tokens: 500 });
    await fetch("https://slack.com/api/chat.postMessage", {
      method: "POST",
      headers: { Authorization: `Bearer ${process.env.SLACK_BOT_TOKEN}`, "Content-Type": "application/json" },
      body: JSON.stringify({ channel: e.channel, text: r.choices[0].message.content }),
    });
  }
});
app.listen(3000);

Set SLACK_BOT_TOKEN + APIVAI_API_KEY, expose the URL to Slack, and @mention the bot.

Polish

Reply in a thread (thread_ts) to keep channels tidy.
Keep short per-thread history for context; trim to control cost.
GPT-5.5 for general Q&A; Claude Sonnet for code.

FAQ

Can a Slack bot use APIVAI? Yes — it's an OpenAI chat call with base URL
https://api.apivai.com/v1; the bot forwards the mention and posts the reply.

Which model? GPT-5.5 for general/multilingual; Claude Sonnet for code-heavy help.

How do I control cost? Trim history and cap max_tokens; APIVAI's per-token price is low.

Get started

Create the Slack app, run the server with your APIVAI key, and @mention the bot. Examples:
APIVAI examples repo.

How to Connect Jan to APIVAI (Cloud Claude & GPT Models)

APIVAI — Wed, 24 Jun 2026 13:47:42 +0000

Connect Jan to APIVAI

Jan is an open-source, offline-first ChatGPT alternative. Besides local models, it
can use remote OpenAI-compatible engines — so you can add APIVAI to chat with cloud Claude and GPT
models at a fraction of list price.

Configure

In Jan, open Settings → Model Providers (or the engine/remote-API settings).
Add an OpenAI-compatible remote engine:
- Base URL / API URL: https://api.apivai.com/v1
- API Key: your APIVAI key
Add the model names you want (e.g. claude-sonnet-4-6, gpt-5.5).
Select an APIVAI model in a new thread.

Confirm models

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

Use a returned name.

Troubleshooting

No response / 401 — recheck the Base URL (/v1) and key.
model_not_found — use an exact name from /v1/models.

Why APIVAI for Jan

Keep local models for offline use and add APIVAI for powerful cloud Claude/GPT models on demand —
one key, a fraction of list price, crypto/USDT/Alipay, no VPN.

FAQ

Can Jan use cloud models via APIVAI? Yes — add an OpenAI-compatible remote engine with base URL
https://api.apivai.com/v1 and your APIVAI key.

Does this replace local models? No — it's an additional remote engine; use whichever per chat.

Which model? Claude Sonnet for depth; GPT-5.5 for fast, cheap replies.

Get started

Add APIVAI as a remote engine in Jan and pick a model from /v1/models. Examples:
APIVAI examples repo.

Build an AI Email Assistant with a Cheap OpenAI-Compatible API

APIVAI — Wed, 24 Jun 2026 09:51:52 +0000

Build an AI email assistant

An AI email assistant can summarize long threads, draft replies in your voice, and triage your
inbox. The model does the language work; APIVAI provides it cheaply over an OpenAI-compatible API
(GPT-5.5 is a great fit for natural, multilingual email).

Core actions

from openai import OpenAI
ai = OpenAI(api_key="YOUR_APIVAI_API_KEY", base_url="https://api.apivai.com/v1")

def summarize(thread):
    r = ai.chat.completions.create(model="gpt-5.5", messages=[
        {"role": "system", "content": "Summarize this email thread in 3 bullets and list any action items."},
        {"role": "user", "content": thread}])
    return r.choices[0].message.content

def draft_reply(thread, intent, tone="friendly and professional"):
    r = ai.chat.completions.create(model="gpt-5.5", messages=[
        {"role": "system", "content": f"Write an email reply that is {tone}. Match the thread's language."},
        {"role": "user", "content": f"Thread:\n{thread}\n\nMy intent: {intent}"}])
    return r.choices[0].message.content

Wire it to your inbox

Gmail/Outlook API or IMAP: fetch messages, run summarize/draft_reply, save drafts.
Triage: classify into categories (urgent / FYI / newsletter) and label accordingly.
Human in the loop: save as a draft for you to review, don't auto-send.

Tips

Put your signature, common facts, and tone guidelines in the system prompt for consistency.
Summarize first for long threads, then draft from the summary to save tokens.
GPT-5.5 handles multilingual email well; a smaller model can do bulk triage cheaply.

FAQ

Which model for email? GPT-5.5 — natural, multilingual, cheap via APIVAI; a smaller model for
bulk triage.

Does APIVAI read my email? No — your code fetches email; only the text you send in the prompt
reaches the model. Keep secrets out of prompts.

Can it auto-send? Better to save drafts for review; auto-send only after you trust it.

Get started

Get an APIVAI key, wire the functions above to your inbox, and start with summaries + drafts.
Examples: APIVAI examples repo.

DEV Community: APIVAI

How to Connect Roo Code to APIVAI (Cheap API for an AI Coding Agent)

Connect Roo Code to APIVAI

Configure the provider

Pick a model

Tips for agent workloads

Troubleshooting

FAQ

Get started

How to Build a Telegram AI Bot with APIVAI (Cheap Claude & GPT)

Build a Telegram AI bot with APIVAI

1. Create the bot

2. The bot (Node.js, long polling — no server needed)

3. Add commands & polish

Pick the model

FAQ

Get started

The Cheapest Way to Use Claude & GPT APIs in 2026

The cheapest way to use Claude & GPT in 2026

Why it works: OpenAI-compatible everywhere

What to look for in a cheap provider

How APIVAI fits

Cut costs further

Get started

How to Use Continue.dev with a Cheap OpenAI-Compatible API

Use Continue.dev with a cheap OpenAI-compatible API

Why it works

Configure Continue

Pick a model that exists

Things to verify

Why APIVAI

Get started

How to Run Whisper Locally for Real-Time Speech-to-Text (faster-whisper)

Run Whisper locally for real-time speech-to-text

Install

Choose a model size

Transcribe in near real time

Hook it to translation (APIVAI)

Performance tips

FAQ

Get started

How to Build a Discord AI Bot with APIVAI (Cheap Claude & GPT)

Build a Discord AI bot with APIVAI

1. Create the bot

2. The bot (discord.js)

Polish

FAQ

Get started

Build an Enterprise Knowledge Base with Open WebUI + APIVAI (RAG)

Build an enterprise knowledge base with Open WebUI + APIVAI

Why this stack

1. Run Open WebUI and connect APIVAI

2. Add your knowledge base

3. Choose the answering model

4. Make answers trustworthy

5. Embeddings note

Example use cases

FAQ

Get started

How to Connect Open WebUI to APIVAI (OpenAI-Compatible)

Connect Open WebUI to APIVAI

1. Run Open WebUI

2. Add APIVAI as an OpenAI connection

3. Pick a model that exists

4. Verify

Troubleshooting

Why APIVAI for Open WebUI

Next

How to Connect Chatbox to APIVAI (Cheap Claude & GPT)

Connect Chatbox to APIVAI

Configure

Confirm available models

Troubleshooting

Why APIVAI for Chatbox

FAQ

Get started

How to Build a Slack AI Bot with APIVAI (Cheap Claude & GPT)

Build a Slack AI bot with APIVAI

1. Create the Slack app

2. The server (Node)