analogaisdk 0.6.0

Minimal Python SDK for AnalogAI completion endpoints

AnalogAI SDK (minimal)

Tiny Python client for the Deepthink OpenAI-compatible chat server (analogai.deepthink.chat). It calls POST {CHAT_BASE_URL}/v1/chat/completions — a stateful endpoint: memory tenancy is keyed by user_id + agent_id (both created server-side on first use), Deepthink memory is retrieved per turn, and messages are ingested in the background.

The return types follow the OpenAI Python SDK schema exactly, so migrating a script is a one-line client swap:

completion = client.generate_completion("hi")
print(completion.choices[0].message.content)
print(completion.usage.prompt_tokens, completion.usage.completion_tokens)

Install

pip install analogaisdk

Setup

Set environment variables in .env:

• CHAT_BASE_URL (default https://cloud.analogai.net:8010; set to http://localhost:8010 for a local server)
• ANALOGAI_API_KEY (optional, if your server requires auth)
• ANALOGAI_MODEL (optional; server default is used when omitted)

Non-streaming

Returns a ChatCompletion matching the OpenAI response shape.

from analogaisdk import AnalogAIClient

client = AnalogAIClient(agent_id="my-agent", user_id="my-user")

completion = client.generate_completion("are humans mortal?")
print(completion.choices[0].message.content)
print("prompt_tokens    =", completion.usage.prompt_tokens)
print("completion_tokens=", completion.usage.completion_tokens)
print("total_tokens     =", completion.usage.total_tokens)

Streaming

Yields ChatCompletionChunk objects. The final chunk carries usage with token counts (the SDK sends stream_options={"include_usage": true} automatically).

for chunk in client.generate_streaming_completion("are humans mortal?"):
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)
    if chunk.usage is not None:
        print()
        print(chunk.usage.prompt_tokens, chunk.usage.completion_tokens)

Working memory

Working memory (a string or list of strings) is appended to the LLM system prompt alongside the Deepthink memory:

client.generate_completion(
    "is Socrates mortal?",
    working_memory=["Socrates is a human"],
)

Explicit messages and model

client.generate_completion(
    messages=[
        {"role": "user", "content": "Say hello"},
    ],
    model="azure:gpt-5.4-mini",  # optional, provider-prefixed
    temperature=0.2,
)

Response schema

• ChatCompletion.id / object / created / model
• ChatCompletion.choices[i].index / .finish_reason
• ChatCompletion.choices[i].message.role / .content / .tool_calls
• ChatCompletion.usage.prompt_tokens / .completion_tokens / .total_tokens
• ChatCompletion.raw — the parsed JSON body from the server, unchanged

For streaming:

• ChatCompletionChunk.choices[i].delta.role / .content / .tool_calls
• ChatCompletionChunk.choices[i].finish_reason
• ChatCompletionChunk.usage — None on intermediate chunks, populated on the final chunk when the server honours stream_options.include_usage.