Verzoeken doen aan DeepSeek-modellen

Werken met DeepSeek in Python

James Chapman

Curriculum Manager, DataCamp

Samenvatting

Een computer die een verzoek naar de Together.ai-API stuurt, met het model en alle data en modelparameters.

Een verzoek maken

from openai import OpenAI

# To send to DeepSeek API: base_url="https://api.deepseek.com"
client = OpenAI(api_key="<TogetherAI API Key>", base_url="https://api.together.xyz/v1")

Verzoeken maken met de openai-bibliotheek
OpenAI: AI-model- en appbouwers (bijv. ChatGPT)
base_url: stuur het verzoek door naar de DeepSeek-provider
api_key: API-authenticatie (zie de documentatie van de provider)

¹ https://platform.deepseek.com/api_keys

Een verzoek maken

from openai import OpenAI
# To send to DeepSeek API: base_url="https://api.deepseek.com"
client = OpenAI(api_key="<TogetherAI API Key>", base_url="https://api.together.xyz/v1")


response = client.chat.completions.create(

    # On DeepSeek's API: model="deepseek-chat"
    model="deepseek-ai/DeepSeek-V3",

    messages=[{"role": "user",
               "content": "In one sentence, what is hallucination in AI?"}]

)


print(response)

De response

ChatCompletion(id='ns1zcjp-zqrih-937de597af0fd643', choices=[Choice(finish_reason='stop', index=0,
logprobs=None, message=ChatCompletionMessage(content='Hallucination in AI refers to the generation of
false or nonsensical information that is presented as factual, often due to limitations in training
data, model biases, or inference errors.', refusal=None, role='assistant', annotations=None, audio=None,
function_call(None), tool_calls=[]), seed=12469585682595789000)], created=1745920244,
model='deepseek-ai/DeepSeek-V3', object='chat.completion', service_tier=None, system_fingerprint=None,
usage=CompletionUsage(completion_tokens=38, prompt_tokens=14, total_tokens=52,
completion_tokens_details=None, prompt_tokens_details=None), prompt=[])

De response

ChatCompletion(id='ns1zcjp-zqrih-937de597af0fd643',
               choices=[Choice(finish_reason='stop', index=0, logprobs=None,
                               message=ChatCompletionMessage(content='Hallucination in AI refers to the
                                   generation of false or nonsensical information that is presented as
                                   factual, often due to limitations in training data, model biases, or
                                   inference errors.',
                               refusal=None, role='assistant', annotations=None, audio=None,
                               function_call=None, tool_calls=[]), seed=12469585682595789000)],
               created=1745920244,
               model='deepseek-ai/DeepSeek-V3',
               object='chat.completion',
               service_tier=None,
               system_fingerprint=None,
               usage=CompletionUsage(completion_tokens=38, prompt_tokens=14, total_tokens=52,
                                     completion_tokens_details=None, prompt_tokens_details=None),
                                     prompt=[])

De response interpreteren

print(response.choices)

[Choice(finish_reason='stop', index=0, logprobs=None, message=ChatCompletionMessage(
content='Hallucination in AI refers to the generation of false or nonsensical information
that is presented as factual, often due to limitations in training data, model biases, or
inference errors.', refusal=None, role='assistant', annotations=None, audio=None,
function_call=None, tool_calls=[]), seed=12469585682595789000)]

De response interpreteren

print(response.choices[0])

Choice(finish_reason='stop', index=0, logprobs=None, message=ChatCompletionMessage(
content='Hallucination in AI refers to the generation of false or nonsensical information
that is presented as factual, often due to limitations in training data, model biases, or
inference errors.', refusal=None, role='assistant', annotations=None, audio=None,
function_call=None, tool_calls=[]), seed=12469585682595789000)

De response interpreteren

print(response.choices[0].message)

ChatCompletionMessage(content='Hallucination in AI refers to the generation of false or
nonsensical information that is presented as factual, often due to limitations in training
data, model biases, or inference errors.', refusal=None, role='assistant', annotations=None,
audio=None, function_call=None, tool_calls=[])

print(response.choices[0].message.content)

Hallucination in AI refers to the generation of false or nonsensical information that is
presented as factual, often due to limitations in training data, model biases, or inference
errors.

API-kosten

Voor DeepSeek:
- Provider (bijv. together.ai)
- Model
- Grotere input + output = hogere kosten

Een tolweg.

¹ https://api-docs.deepseek.com/quick_start/pricing

Laten we oefenen!

Werken met DeepSeek in Python