curl -X POST https://api.core.today/llm/openai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer cdt_your_api_key" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}],
    "max_tokens": 1000,
    "temperature": 0.7
  }'

GPT-5 / O-Series (추론 모델)

curl -X POST https://api.core.today/llm/openai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer cdt_your_api_key" \
  -d '{
    "model": "gpt-5",
    "messages": [{"role": "user", "content": "Explain quantum computing"}],
    "max_completion_tokens": 16000,
    "reasoning_effort": "medium"
  }'

GPT-5 전용 파라미터

max_completion_tokens - 최대 출력 토큰 (max_tokens 대신 사용)
reasoning_effort - 추론 수준: minimal, low, medium, high

GPT-5 시리즈

모델	Input	Output
gpt-5	0.0025	0.0200
gpt-5-mini	0.0005	0.0040
gpt-5-nano	0.0001	0.0008
gpt-5-pro	0.0300	0.2400

GPT-4.1 / GPT-4o 시리즈

모델	Input	Output
gpt-4.1	0.0040	0.0160
gpt-4.1-mini	0.0008	0.0032
gpt-4.1-nano	0.0002	0.0008
gpt-4o	0.0050	0.0200
gpt-4o-mini	0.0003	0.0012

O-Series (Reasoning)

모델	Input	Output
o1	0.0300	0.1200
o1-pro	0.3000	1.2000
o3	0.0040	0.0160
o3-pro	0.0400	0.1600
o4-mini	0.0022	0.0088

Anthropic (Claude)

curl -X POST https://api.core.today/llm/anthropic/v1/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer cdt_your_api_key" \
  -d '{
    "model": "claude-sonnet-4",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Explain quantum computing simply."}
    ]
  }'

Claude Opus 시리즈

모델	Input	Output
claude-opus-4.5	0.0100	0.0500
claude-opus-4.1	0.0300	0.1500
claude-opus-4	0.0300	0.1500
claude-3-opus	0.0300	0.1500

Claude Sonnet 시리즈

모델	Input	Output
claude-sonnet-4.5	0.0060	0.0300
claude-sonnet-4	0.0060	0.0300
claude-3.7-sonnet	0.0060	0.0300
claude-3.5-sonnet	0.0060	0.0300

Claude Haiku 시리즈

모델	Input	Output
claude-haiku-4.5	0.0020	0.0100
claude-3.5-haiku	0.0016	0.0080
claude-3-haiku	0.0005	0.0025

Google (Gemini)

curl -X POST "https://api.core.today/llm/gemini/v1beta/models/gemini-2.5-pro:generateContent" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer cdt_your_api_key" \
  -d '{
    "contents": [
      {
        "parts": [{"text": "Write a haiku about programming"}]
      }
    ]
  }'

Gemini 모델

모델	Input	Output
gemini-2.5-pro	0.0025	0.0200
gemini-2.5-flash	0.0006	0.0050
gemini-2.5-flash-lite	0.0002	0.0008
gemini-2.0-flash	0.0002	0.0008
gemini-2.0-flash-live	0.0007	0.0030

참고: gemini-2.5-pro는 200,000 토큰 초과 시 longcontext 가격 적용 (Input: 0.0050, Output: 0.0300)

스트리밍 응답

실시간으로 응답을 받으려면 stream: true를 추가하세요:

curl -X POST https://api.core.today/llm/openai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer cdt_your_api_key" \
  -d '{
    "model": "gpt-5",
    "messages": [{"role": "user", "content": "Tell me a long story"}],
    "stream": true
  }'

비용 계산 예시

1,000 토큰 입력, 500 토큰 출력 기준:

모델	계산	총 비용
gpt-4o-mini	0.3 + 0.6	0.9 크레딧
gpt-5	2.5 + 10.0	12.5 크레딧
claude-3-haiku	0.5 + 1.25	1.75 크레딧
claude-sonnet-4	6.0 + 15.0	21.0 크레딧
gemini-2.0-flash	0.2 + 0.4	0.6 크레딧