> For the complete documentation index, see [llms.txt](https://langtech-bsc.gitbook.io/applied-intelligence/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://langtech-bsc.gitbook.io/applied-intelligence/which-model.md).

# Which model should I use?

Match your task to the right model in under a minute.

| I want to…                                                  | Use                                                                                                                                                                                                                            | Notes                                                                      |
| ----------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------- |
| Build an assistant that calls my APIs and tools             | [ALIA-40b-fc-2607](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/alia-40b-fc.md)                                                                                                                     | Our flagship. Multi-GPU in BF16, or a single 48 GB GPU with the 4-bit GGUF |
| …the same, but on a single small GPU                        | [Prelude-9b](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/prelude-9b.md) or [Salamandra-7b-fc-evo](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/salamandra-7b-fc-evo.md) | Coming soon                                                                |
| Answer questions over long documents                        | [ALIA-40b-fc-2607](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/alia-40b-fc.md)                                                                                                                     | 160k-token context window                                                  |
| Transcribe Catalan audio for people to read                 | [whisper-large-v3-ca-punctuated-3370h](https://langtech-bsc.gitbook.io/applied-intelligence/models/speech-models/whisper-large-v3-ca-punctuated.md)                                                                            | Punctuation and capitalization included                                    |
| Benchmark ASR, or transcribe mixed Catalan/Spanish audio    | [stt\_ca-es\_conformer\_transducer\_large](https://langtech-bsc.gitbook.io/applied-intelligence/models/speech-models/conformer-transducer-ca-es.md)                                                                            | Lowest WER, small and fast                                                 |
| Generate Central Catalan speech (also commercially)         | [matxa-tts-v2-ca-central-graphemes](https://langtech-bsc.gitbook.io/applied-intelligence/models/speech-models/matxa-tts-central.md)                                                                                            | GPL-3.0. Pair it with the WaveNext vocoder                                 |
| Generate speech in a regional accent (Valencian, Balearic…) | [matxa-tts-v2-ca-multiaccent-graphemes](https://langtech-bsc.gitbook.io/applied-intelligence/models/speech-models/matxa-tts-multiaccent.md)                                                                                    | 16 voices. Non-commercial license                                          |
| Translate with the best quality, glossaries or HTML/XML     | [salamandraTA-7b-instruct](https://langtech-bsc.gitbook.io/applied-intelligence/models/machine-translation/salamandrata-7b-instruct.md)                                                                                        | 40 languages. GPL-3.0                                                      |
| Translate on a CPU or under a permissive license            | [salamandraTA-2b-instruct](https://langtech-bsc.gitbook.io/applied-intelligence/models/machine-translation/salamandrata-2b-instruct.md)                                                                                        | 35 languages. Apache 2.0. 1.5 GB GGUF                                      |
| Route, classify, score or moderate Catalan text             | [CatDecide-1](https://langtech-bsc.gitbook.io/applied-intelligence/models/decision-models/catdecide-1.md)                                                                                                                      | Calibrated confidence, no text generation. Coming soon                     |

## Quick decision tree

{% stepper %}
{% step %}

### Is the output a label, a yes/no or a score?

Use a **decision model**: [CatDecide-1](https://langtech-bsc.gitbook.io/applied-intelligence/models/decision-models/catdecide-1.md). It's faster and gives a calibrated confidence for every answer.
{% endstep %}

{% step %}

### Is the input or output audio?

Speech to text: use [Whisper](https://langtech-bsc.gitbook.io/applied-intelligence/models/speech-models/whisper-large-v3-ca-punctuated.md) for readable transcripts and the [Conformer](https://langtech-bsc.gitbook.io/applied-intelligence/models/speech-models/conformer-transducer-ca-es.md) for scoring or mixed-language audio. Text to speech: use [Matxa-TTS](https://langtech-bsc.gitbook.io/applied-intelligence/models/speech-models.md).
{% endstep %}

{% step %}

### Is the task purely translation?

Use [SalamandraTA](https://langtech-bsc.gitbook.io/applied-intelligence/models/machine-translation.md). It's trained specifically for translation, including low-resource languages of Spain such as Asturian, Aragonese and Aranese.
{% endstep %}

{% step %}

### Everything else

Chat, RAG, tool use, agents: use a [text model](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models.md).
{% endstep %}
{% endstepper %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://langtech-bsc.gitbook.io/applied-intelligence/which-model.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
