Access DeepSeek V4, DeepSeek R1, MiMo, TTS, and ASR models through a single unified endpoint — up to 82% cheaper than major API marketplaces. Starting at just $0.19 / 277₩ per million tokens.
curl https://nexusaipro.xyz/v1/chat/completions \
-H "Authorization: Bearer sk-nx-YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-r1",
"messages": [{"role": "user", "content": "Hello!"}]
}'Up to 90% cheaper than GPT-4o. Powerful models at a fraction of the cost, with transparent per-token pricing.
Fully OpenAI- and Anthropic-compatible API. Works with LangChain, LlamaIndex, OpenAI SDK, Claude Code, and any HTTP client.
No Chinese phone number or payment method needed. Pay with credit card, PayPal, or Apple Pay via Paddle.
Access DeepSeek R1 reasoning model for complex math, coding, and logic tasks at unbeatable prices.
Text-to-Speech and Speech Recognition — all through one API.
Invite friends and earn 10% of their first top-up. Share your unique referral code and start earning.
Pay only for what you use. Prices shown in USD / KRW per 1M tokens.
Top-tier reasoning and coding
Ultra-fast, cost-effective inference
Deep thinking model — math, code, logic
Fast and efficient for everyday tasks
Advanced reasoning and complex tasks
Peak hours for DeepSeek v4 models only: 09:00–12:00 & 14:00–18:00 (Beijing Time, UTC+8). DeepSeek R1 is not affected.
Compared with public list prices of large resellers (e.g. AIHubMix), August 2026. Per 1M tokens.
| Model | Typical reseller | NexusAI | You save |
|---|---|---|---|
| DeepSeek V4-Flash — input | $0.55 | $0.19 | 65% |
| DeepSeek V4-Flash — output | $2.20 | $0.39 | 82% |
| DeepSeek V4-Pro — input | $1.69 | $0.65 | 62% |
| DeepSeek V4-Pro — output | $3.38 | $1.30 | 62% |
Official APIs only — no quantized or degraded models.
Text-to-Speech and Speech Recognition. Prices in USD / KRW per hour of audio.
Text-to-Speech synthesis
Top up anytime. Credits never expire. Start with a $1 trial.
from openai import OpenAI
client = OpenAI(
api_key="sk-nx-YOUR_KEY",
base_url="https://nexusaipro.xyz/v1"
)
response = client.chat.completions.create(
model="deepseek-r1",
messages=[{"role": "user", "content": "Hi!"}]
)
print(response.choices[0].message.content)from openai import OpenAI
client = OpenAI(
api_key="sk-nx-YOUR_KEY",
base_url="https://nexusaipro.xyz/v1"
)
response = client.audio.speech.create(
model="mimo-v2.5-tts",
voice="alloy",
input="Hello, welcome to NexusAI!"
)
response.stream_to_file("output.mp3")import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: "sk-nx-YOUR_KEY",
baseURL: "https://nexusaipro.xyz",
});
const msg = await client.messages.create({
model: "deepseek-v4-pro",
max_tokens: 1024,
messages: [{ role: "user", content: "Hi!" }],
});
console.log(msg.content[0].text);
// Claude Code (terminal):
// export ANTHROPIC_BASE_URL="https://nexusaipro.xyz/v1"
// export ANTHROPIC_API_KEY="sk-nx-YOUR_KEY"