Vynex API

One OpenAI-compatible endpoint for every frontier model.

Call GPT, Claude, Gemini, DeepSeek, Qwen and GLM-5.2 from a single /v1 endpoint. GLM-5.2 at $0.91 / $2.86 per 1M tokens (volume tier) — 65% of Z.ai official pricing. Keep your existing OpenAI SDK — just change the base_url.

Website · Docs · Pricing · Get API Key

---

Why Vynex?

• One key, one endpoint. GPT-5.x, Claude Opus/Sonnet/Haiku, Gemini 3.x, DeepSeek, Qwen, GLM — all behind a single OpenAI-compatible /v1.
• Zero migration. If your code works with the OpenAI SDK, it works with Vynex. Change the base_url and you're done.
• Swap models in one line. Switch providers by changing the model field — no re-integration.
• Transparent per-token pricing. See Pricing.
• GLM-5.2: cheapest frontier model. $0.91 input / $2.86 output per 1M tokens at the volume tier — 65% of Z.ai official, cheaper than DeepInfra. MIT-licensed, 34% hallucination rate (beats Claude Sonnet 4.5). Same key, same endpoint.
• Pay with USDT — no foreign credit card needed. Top up with USDT (TRC20/ERC20) if OpenAI/Anthropic cards get declined in your region. Cards and bank transfer also supported. Built for devs in MENA, SEA, Russia, Iran and LatAm.
• Streaming, function/tool calling, structured output — the OpenAI params you already know.

---

Integrations

Step-by-step guides for popular tools (OpenAI-compatible, zero SDK changes):

Tool	Guide
Cursor	integrations/cursor.md
Cline (VS Code)	integrations/cline.md
LangChain	integrations/langchain.md

Quick start

1. Get an API key

Create a key at llm-api.vynexcloud.com/register.

2. Point your OpenAI client at Vynex

Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-...",
    base_url="https://llm-api.vynexcloud.com/v1",
)

response = client.chat.completions.create(
    model="gpt-5.4",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

Node.js / TypeScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.VYNEX_API_KEY,
  baseURL: "https://llm-api.vynexcloud.com/v1",
});

const res = await client.chat.completions.create({
  model: "claude-sonnet-4-6",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(res.choices[0].message.content);

cURL

curl https://llm-api.vynexcloud.com/v1/chat/completions \
  -H "Authorization: Bearer $VYNEX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

That's it. To switch from GPT to Claude, Gemini, or any other model, change only the model field.

---

Supported models

Family	Models
OpenAI	gpt-5.5, gpt-5.4, gpt-5.4-mini, gpt-5.4-nano
Anthropic	claude-opus-4-8, claude-sonnet-4-6, claude-haiku-4-5-20251001
Google	gemini-3.1-pro-preview, gemini-3-pro-preview, gemini-2.5-pro, gemini-2.5-flash
Open-source	deepseek/deepseek-v3.2, qwen3.5-plus, glm-5.2

List the live set anytime: curl https://llm-api.vynexcloud.com/v1/models -H "Authorization: Bearer $VYNEX_API_KEY". New models are added frequently.

---

Use with your tools

Vynex works anywhere a custom OpenAI-compatible base_url is accepted:

Tool	base_url	model
Cursor	https://llm-api.vynexcloud.com/v1	any supported
Cline	https://llm-api.vynexcloud.com/v1	any supported
Continue	https://llm-api.vynexcloud.com/v1	any supported
Open WebUI	https://llm-api.vynexcloud.com/v1	any supported
LobeChat	https://llm-api.vynexcloud.com/v1	any supported
LangChain	OpenAI(base_url="https://llm-api.vynexcloud.com/v1")	any supported
LlamaIndex	same	any supported

See examples/ for ready-to-run snippets.

---

Features

• Streaming — standard SSE stream: true
• Function / tool calling — OpenAI tools API
• Embeddings — /v1/embeddings
• Structured output — response_format

---

Error handling & limits

Standard OpenAI-style HTTP status codes apply:

Status	Meaning
401	Invalid or missing API key
429	Rate limit exceeded — retry with backoff
5xx	Upstream error — retry or fall back to another model

Tip: build a fallback chain across model families for resilience.

---

Security & privacy

• API keys are encrypted at rest.
• Usage is metered per-token for billing.
• See our Privacy Policy and Terms.

---

Community

• 🐛 Report issues — bugs, feature requests, model requests
• 📖 Documentation

Contributing

PRs welcome — especially example configs for new tools and frameworks. See CONTRIBUTING.md.

License

MIT

Get your API key →

Built with ❤️ by the Vynex team.