Minimax AI

Minimax's language models with SSE streaming support. Requires a group_id in addition to an API key.

Install

No extra dependencies needed — MinimaxLLM uses the built-in httpx client.

Setup

export MINIMAX_API_KEY=your-api-key
export MINIMAX_GROUP_ID=your-group-id

Usage

from synapsekit.llm.minimax import MinimaxLLM
from synapsekit import LLMConfig
import os

config = LLMConfig(
    model="abab6.5s-chat",
    api_key=os.environ["MINIMAX_API_KEY"],
    provider="minimax",
)

llm = MinimaxLLM(
    config,
    group_id=os.environ["MINIMAX_GROUP_ID"],
)

# Streaming
async for token in llm.stream("Explain SSE streaming"):
    print(token, end="")

# Generate
response = await llm.generate("What is Minimax AI?")

Available models

Model	Notes
abab6.5s-chat	Fast, efficient
abab6.5-chat	Balanced quality
abab5.5s-chat	Previous generation

Auto-detection

The RAG facade auto-detects Minimax for minimax-* model prefixes:

from synapsekit import RAG
import os

rag = RAG(
    model="minimax-abab6.5s",
    api_key=os.environ["MINIMAX_API_KEY"],
    provider="minimax",
)
rag.add("Your document text here")
answer = rag.ask_sync("Summarize this.")

tip

Set MINIMAX_GROUP_ID as an environment variable — it's required for all Minimax API calls and is automatically read from the environment.