Same cutting-edge AI models — DeepSeek-V3, DeepSeek-R1, and more. Enterprise-grade API with developer-friendly pricing. No minimums. No commitments.
All prices per 1M tokens. OpenRouter-compatible API. No markup on cached tokens.
| Model | Input | Output | OpenAI Equivalent | Save |
|---|---|---|---|---|
DeepSeek-V3 |
$0.44 | $0.44 | $2.50 (GPT-4o mini) | 83% cheaper |
DeepSeek-R1 |
$2.62 | $2.62 | $15.00 (o1-mini) | 83% cheaper |
DeepSeek-V4 Flash |
$0.24 | $0.80 | $10.00 (GPT-4o) | 92% cheaper |
DeepSeek-R1 0328 |
$2.62 | $2.62 | $15.00 (o1-mini) | 83% cheaper |
DeepSeek-V3 0324 |
$0.54 | $0.54 | $2.50 (GPT-4o mini) | 78% cheaper |
Pay as you go. No subscriptions. Every dollar gives you 500,000 quota units.
All models at a fraction of OpenAI prices. Change base_url to https://api.centurygold.site and you are ready.
| Model | Equivalent To | Price | Speed | Best For |
|---|---|---|---|---|
| DeepSeek-V3 | GPT-4o | 1/10 GPT-4o price | ⚡ Fast | Coding, logical reasoning, general tasks |
| DeepSeek-R1 | OpenAI o1 | 1/10 o1 price | 🐢 Deep thinking | Complex math, science, hard problems |
| DeepSeek-V4-Flash | GPT-4o-mini | Cheaper than GPT-4o-mini | ⚡⚡ Very fast | Chat, quick Q&A, lightweight tasks |
| Qwen 3.5-122B | Llama-3-70B | $1.00 / M tokens | ⚡ Fast | Long context, Chinese & English |
| Qwen 3.5 (35B/27B/9B) | Other open models | $0.12-0.24 / M tokens | ⚡⚡ Very fast | Budget-friendly, prototyping, batch jobs |
All prices are per million input tokens. Output tokens billed at same rate.
Built for developers who want quality AI without the premium price tag.
Use your existing code — just change the base URL. One line change.
Register, create a key, and start calling. No approval process.
Track every request. See spend in real-time. Set spending limits.
Singapore-based servers with low latency across Asia and the US.
Powered by Paddle. Pay with credit card, PayPal, or Apple Pay.
Start with $5. No monthly commitments. No hidden fees.
Get started with $5 in credits. No subscription required.
⚡ Get Credits Now