Add MiniMax RAG cookbook notebook

[octo-patch]

Summary

Add a new cookbook notebook demonstrating how to use MiniMax large language models with Haystack via the OpenAI-compatible API.

What's included

• notebooks/rag_with_minimax.ipynb — Cookbook notebook covering:
  • Basic chat with MiniMax using OpenAIChatGenerator
  • Full RAG pipeline with InMemoryDocumentStore, sentence-transformer embeddings, and MiniMax as the generator
  • Model comparison between MiniMax-M2.7 (204K context) and MiniMax-M2.7-highspeed (speed-optimized)
• index.toml — Added entry for the new notebook
• tests/test_rag_with_minimax.py — 34 tests (31 unit + 3 integration)

Why MiniMax?

MiniMax provides models with a 204K context window accessible via an OpenAI-compatible API (https://api.minimax.io/v1). This means no extra integration packages are needed — Haystack's built-in OpenAIChatGenerator works out of the box by setting api_base_url.

Key configuration

from haystack.components.generators.chat import OpenAIChatGenerator
from haystack.utils import Secret

generator = OpenAIChatGenerator(
    api_key=Secret.from_env_var("MINIMAX_API_KEY"),
    model="MiniMax-M2.7",
    api_base_url="https://api.minimax.io/v1",
)

Test plan

• 31 unit tests pass (notebook structure, config validation, index.toml)
• 3 integration tests pass (MiniMax API chat completion, Haystack generator)
• scripts/verify_index.py passes

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[bilgeyucel]

Hi @octo-patch, thanks for the PR! I see that you use this notebook to showcase how to use MiniMax models with Haystack. Instead of adding a new cookbook, would you like to add a new integration page to haystack-integrations with the code examples and information you have? I see that we are missing that.
Let me know if you have any questions!