​
Models

For Completions, we currently support the following models:

  • cygnet-llama-3

The model name cygnet will point to our latest model.

​
Messages

To interact with our model, you will need to format your prompt as a series of messages. Each message has a role and content. Different roles are treated differently by the model. The available roles are:

  • system (optional): Used to provide instructions, guidelines, and context to the model regarding the ensuing conversation between user and model
  • user: Used for messages sent by the user to the model
  • data: See the guide on Data elements for an explanation of this role
  • assistant: Used for messages sent by our model to the user

In our Python client, messages are represented as follows:

messages=[
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "What day is after Monday?"},
    {"role": "assistant", "content": "The day after Monday is Tuesday."}
]

​
Completions

Below we provide examples of usage of our Completions endpoint, for non-streaming, streaming, async non-streaming, and async with streaming.

​
Non-streaming

import os
from gray_swan import GraySwan
GRAYSWAN_API_KEY = os.environ.get("GRAYSWAN_API_KEY")

client = GraySwan(
    api_key=GRAYSWAN_API_KEY,
)

completion_create_response = client.chat.completion.create(
    messages=[{"role": "system", "content": "You are a helpful assistant."},
              {"role": "user", "content": "What is a large language model?"}],
    model="cygnet",
)

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

​
Streaming

completion_create_response = client.chat.completion.create(
    messages=[{"role": "system", "content": "You are a helpful assistant."},
              {"role": "user", "content": "What is a large language model?"}],
    model="cygnet",
    stream=True
)
for r in completion_create_response:
    delta_content = r.choices[0].delta.content
    print(delta_content, end="")

​
Async without streaming

import os
from gray_swan import AsyncGraySwan
GRAYSWAN_API_KEY = os.environ.get("GRAYSWAN_API_KEY")

client = AsyncGraySwan(
    api_key=GRAYSWAN_API_KEY,
)

completion_create_response = await client.chat.completion.create(
    messages=[{"role": "system", "content": "You are a helpful assistant."},
              {"role": "user", "content": "What is a large language model?"}],
    model="cygnet",
    stream=True
)
for r in completion_create_response:
    delta_content = r.choices[0].delta.content
    print(delta_content, end="")

​
Async with streaming

import os
from gray_swan import AsyncGraySwan
GRAYSWAN_API_KEY = os.environ.get("GRAYSWAN_API_KEY")

client = AsyncGraySwan(
    api_key=GRAYSWAN_API_KEY,
)
completion_create_response = client.chat.completion.create(
    messages=[{"role": "system", "content": "You are a helpful assistant."},
              {"role": "user", "content": "What is a large language model?"}],
    model="cygnet",
    stream=True
)
async for r in completion_create_response:
    delta_content = r.choices[0].delta.content
    print(delta_content, end="")
QuickstartData elements