DeepSeek Series

Bailian-integrated DeepSeek V4 flagship reasoning model, suited for complex math, long routing decisions, and Code Agent tasks.

Bailian-integrated DeepSeek V4 high-speed version. Low latency, high throughput, suited for online Q&A, customer service, and high-concurrency gateway scenarios.

Reasoning-enhanced model that analyzes complex problems via chain-of-thought reasoning. Outstanding performance in math, coding, and logical reasoning tasks.

Latest-generation general model using MoE architecture, comprehensively improved in coding, math, and general capabilities. Extremely cost-efficient.

from openai import OpenAI

client = OpenAI(
    api_key="sk-air-your-key",
    base_url="https://nexusflow.vip/v1",
)

# Use DeepSeek R1 for reasoning
response = client.chat.completions.create(
    model="deepseek-r1",
    messages=[
        {"role": "user", "content": "Analyze step by step: If A > B, B > C, what is the relationship between A and C?"}
    ],
    max_tokens=2048,
)

# R1 model displays the chain-of-thought process
print(response.choices[0].message.content)