LLM-guide: which model should i use when?
Which language model fits which task? Recommendations on Auto, Claude, GPT, Gemini, reasoning and the open-weight models in MAIA.
MAIA lets you choose between many language models. For most questions you don't need to: with “Auto”, MAIA picks a suitable model for each request. This guide helps when you want to choose a model yourself.
As of October 2026, release 1.25. For the full list of models, see Model Selection Reference. You choose the model in the answer settings under “AI model”.
In short
Section titled “In short”| You want to … | Recommendation |
|---|---|
| ask everyday questions and search your documents | “Auto” |
| get thorough everyday answers, quick and well written | Claude Sonnet 5.5 |
| do in-depth research, complex analyses or multi-step tasks | Claude Opus 5.5, with effort “High” or “Maximum” for very hard questions |
| get quick, short answers to simple questions | GPT-4.1 mini or GPT-4.1 nano |
| use a model with publicly available weights, operated by a European provider | one of the open-weight models, for example Mistral 4 |
“Auto”: the default
Section titled ““Auto”: the default”With “Auto”, MAIA looks at each request, meaning your question, the selected files and the effort level, and picks a suitable model from a small set tested for “Auto”. Within a chat, the model can change from question to question. Your message shows which model MAIA picked.
“Auto” only uses new models once we've tested them thoroughly. We adjust which models belong to that set over time.
The model families
Section titled “The model families”Claude (Anthropic)
Section titled “Claude (Anthropic)”- Claude Opus 5.5 and Opus 5: For the most demanding tasks, such as in-depth research, complex analyses and tasks with many steps. Opus 5.5 is the newer model.
- Claude Sonnet 5.5: A strong everyday all-rounder, thorough and still quick.
All three use reasoning in MAIA.
GPT (OpenAI)
Section titled “GPT (OpenAI)”At OpenAI, the names Sol, Terra and Luna stand for performance tiers: Sol is the stronger variant, Luna the faster one, and Terra sits in between. All GPT-5.6 and GPT-6 models always use reasoning.
- GPT-6 Sol: For demanding tasks such as research and analyses with many steps. According to OpenAI, it makes far fewer mistakes than GPT-5.6 Sol.
- GPT-6 Luna: Fast, for clearly defined tasks, for example summarizing documents, extracting information or answering quick questions.
- GPT-5.6 Sol: The strongest model in the GPT-5.6 series, for difficult, multi-step questions.
- GPT-5.6 Terra: Balances quality and speed, good for everyday tasks and document analysis.
- GPT-5.6 Luna: The fastest model in the GPT-5.6 series. For questions that need a lot of text at once, it performs worse than Sol.
- GPT-4.1, GPT-4.1 mini and GPT-4.1 nano: Work without reasoning and therefore answer quickly. The smaller mini and nano variants suit short, simple questions.
Gemini (Google)
Section titled “Gemini (Google)”- Gemini 2.5 Pro: Always uses reasoning. Google describes it as a model for complex questions and for analyzing large datasets and long documents.
- Gemini 3.5 Flash: Google's fast workhorse model, good for tasks with several steps. Uses reasoning in MAIA.
Open-weight models
Section titled “Open-weight models”Gemma 4, Apertus 70B, DeepSeek V4.1 Flash, GLM 5.3, Mistral 4 and Qwen 3.8 are open-weight models: their model weights are publicly available. In MAIA, they're operated by a European provider.
How thorough should MAIA be?
Section titled “How thorough should MAIA be?”Many models think about your request internally before answering (“reasoning”). How much MAIA researches and reasons is set with the “Effort per answer” in the answer settings:
- Low: short, everyday questions and quick rewrites.
- Standard: everyday work such as summarizing, drafting and asking questions about documents.
- High: comparing several documents or drafting longer material.
- Maximum: complex questions with many sources. Takes much longer.
Higher effort pays off for complex questions, for example when MAIA should compare several documents, spot contradictions or work something out step by step. See Answer settings.
Some models have a usage allowance per person and model that depends on your licence and resets regularly. See Model Selection Reference.
- Stick with “Auto” if you're unsure. It's the best choice for most questions.
- Compare for important questions. Ask the same question with two models and compare the answers.
- Your choice applies per chat. A model you pick applies in that chat until you change it. If you pick it for the first message of a new chat, it also applies to your next new chats.
- For especially precise answers, also switch on “Verify answers”.
Questions or feedback? Email us at support@getmaia.ai.