Deepseek
Deepseek
DeepSeek의 모든 모델을 지원해요. completion 요청 시 deepseek/ 접두사로 설정하기만 하면 돼요.
출처: 문서
본문
API 키
# env variable
os.environ['DEEPSEEK_API_KEY']
샘플 사용법 (Sample Usage)
from litellm import completion
import os
os.environ['DEEPSEEK_API_KEY'] = ""
response = completion(
model="deepseek/deepseek-chat",
messages=[
{"role": "user", "content": "hello from litellm"}
],
)
print(response)
샘플 사용법 - 스트리밍 (Streaming)
from litellm import completion
import os
os.environ['DEEPSEEK_API_KEY'] = ""
response = completion(
model="deepseek/deepseek-chat",
messages=[
{"role": "user", "content": "hello from litellm"}
],
stream=True
)
for chunk in response:
print(chunk)
지원 모델 - 모든 Deepseek 모델 지원!
| 모델 이름 | 함수 호출 |
|---|---|
| deepseek-chat | completion(model="deepseek/deepseek-chat", messages) |
| deepseek-coder | completion(model="deepseek/deepseek-coder", messages) |
| deepseek-flash | completion(model="deepseek/deepseek-flash", messages) |
| deepseek-v4-pro | completion(model="deepseek/deepseek-v4-pro", messages) |
Reasoning 모델
| 모델 이름 | 함수 호출 |
|---|---|
| deepseek-reasoner | completion(model="deepseek/deepseek-reasoner", messages) |
Thinking / Reasoning 모드
DeepSeek reasoner 모델의 thinking 모드는 thinking 또는 reasoning_effort 파라미터로 활성화할 수 있어요.
thinking 파라미터:
from litellm import completion
import os
os.environ['DEEPSEEK_API_KEY'] = ""
resp = completion(
model="deepseek/deepseek-reasoner",
messages=[{"role": "user", "content": "What is 2+2?"}],
thinking={"type": "enabled"},
)
print(resp.choices[0].message.reasoning_content) # Model's reasoning
print(resp.choices[0].message.content) # Final answer
reasoning_effort 파라미터:
from litellm import completion
import os
os.environ['DEEPSEEK_API_KEY'] = ""
resp = completion(
model="deepseek/deepseek-reasoner",
messages=[{"role": "user", "content": "What is 2+2?"}],
reasoning_effort="medium", # low, medium, high all map to thinking enabled
)
print(resp.choices[0].message.reasoning_content) # Model's reasoning
print(resp.choices[0].message.content) # Final answer
DeepSeek은
{"type": "enabled"}만 지원해요 — Anthropic과 달리budget_tokens를 지원하지 않아요."none"이 아닌 모든 reasoning_effort 값이 thinking 모드를 활성화해요.
기본 사용법
SDK:
from litellm import completion
import os
os.environ['DEEPSEEK_API_KEY'] = ""
resp = completion(
model="deepseek/deepseek-reasoner",
messages=[{"role": "user", "content": "Tell me a joke."}],
)
print(resp.choices[0].message.reasoning_content)
Proxy:
model_list:
- model_name: deepseek-reasoner
litellm_params:
model: deepseek/deepseek-reasoner
api_key: os.environ/DEEPSEEK_API_KEY
litellm --config /path/to/config.yaml
curl -L -X POST 'http://0.0.0.0:4000/v1/chat/completions' \
-H 'Content-Type: application/json' \
-H "Authorization: Bearer ***" \
-d '{
"model": "deepseek-reasoner",
"messages": [
{
"role": "user",
"content": [
{
"type": "text",
"text": "Hi, how are you ?"
}
]
}
]
}'