Compatible with OpenAI and Anthropic API formats, zero code changes.
Claude, GPT full model series. Stable, fast, transparent pricing.
50% off official pricing. No minimum spend. Pay per token, only for what you use.
Compatible with OpenAI and Anthropic API formats. Just change base_url and api_key.
Full SSE streaming support, real-time token counting, millisecond-level latency.
from openai import OpenAI
client = OpenAI(
base_url="http://47.253.249.242/v1",
api_key="sk-yq-your-key-here"
)
response = client.chat.completions.create(
model="claude-sonnet-4-20250514",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)| Model | Input (USD/MTok) | Output (USD/MTok) | vs Official |
|---|---|---|---|
| Claude Opus 4 | 7.50 | 37.50 | 50% off |
| Claude Sonnet 4 | 1.50 | 7.50 | 50% off |
| Claude Haiku 3.5 | 0.40 | 2.00 | 50% off |
| GPT-4o | 1.25 | 5.00 | 50% off |