Skip to documentation
deepthroat
DocsAPI reference
DEEPTHROAT DEVELOPER GUIDE

Chat Completions

Send messages with OpenAI-compatible SDKs or a plain HTTP request.

On this pageRequestResponseGeneration parameters

Request

POST/v1/chat/completions
from openai import OpenAI as openai

client = openai(
    api_key="YOUR_API_KEY",
    base_url="https://api.cheapai.gay/v1"
)

reply = client.chat.completions.create(
    model="gpt-6.1-sol",
    messages=[{"role": "user", "content": "Hello, deepthroat."}]
)
print(reply.choices[0].message.content)

Response

json
{
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello!"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 3,
    "total_tokens": 15
  }
}

Read text from choices[0].message.content. A tool request appears in message.tool_calls. Reasoning, when supplied by the model, appears in message.reasoning_content.

Generation parameters

ParameterPurpose
modelExact model ID; required
messagesConversation as role/content messages; required
streamSet true for server-sent events
max_tokensRequested output-token limit
temperatureSampling preference where supported
reasoning_effortA reasoning level accepted by the model
tools / tool_choiceFunction definitions and tool selection

Parameter support varies by model. Read Reasoning controls and Tool calling for examples.