> 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/models/text-models.md).

# Text (LLMs)

Instruction-tuned large language models for assistants, function calling and agentic workflows.

These are the language models we recommend for building assistants, retrieval-augmented generation (RAG), tool use and agents. Each one is trained or post-trained at BSC on the MareNostrum 5 supercomputer. ALIA-40b-fc-2607 is released under the **Apache 2.0** license; licenses for the upcoming models will be announced at release.

## Recommended models

| Model                                                                                                                       | Size | Best for                                                            | Context     | Status                                             |
| --------------------------------------------------------------------------------------------------------------------------- | ---- | ------------------------------------------------------------------- | ----------- | -------------------------------------------------- |
| [**ALIA-40b-fc-2607**](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/alia-40b-fc.md)              | 40B  | Our flagship: reliable function calling and multilingual assistants | 160k tokens | <mark style="color:green;">**Available**</mark>    |
| [**Salamandra-7b-fc-evo**](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/salamandra-7b-fc-evo.md) | 7B   | A lighter function-calling model for the languages of Spain         | —           | <mark style="color:orange;">**Coming soon**</mark> |
| [**Prelude-9b**](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/prelude-9b.md)                     | 9B   | Agentic tool use with explicit reasoning, at a small size           | —           | <mark style="color:orange;">**Coming soon**</mark> |

{% hint style="info" %}
**How to read the names.** `fc` means the model is tuned for **function calling**. The four-digit suffix is the release date as `YYMM`: `2607` is July 2026 and `2610` is October 2026.
{% endhint %}

## Choosing between them

* **Need the strongest model and have a multi-GPU server?** Use [ALIA-40b-fc-2607](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/alia-40b-fc.md). For a single 48 GB GPU, use its 4-bit GGUF build (23 GB of weights).
* **Need something that fits on one consumer GPU?** Watch [Prelude-9b](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/prelude-9b.md) and [Salamandra-7b-fc-evo](https://langtech-bsc.gitbook.io/applied-intelligence/models/text-models/salamandra-7b-fc-evo.md), both coming soon.
* **Only translating text?** A general LLM is the wrong tool. Use [SalamandraTA](https://langtech-bsc.gitbook.io/applied-intelligence/models/machine-translation.md).
* **Only classifying or routing text?** You don't need text generation at all. Use [CatDecide-1](https://langtech-bsc.gitbook.io/applied-intelligence/models/decision-models/catdecide-1.md).

{% hint style="warning" %}
None of these LLMs has gone through safety or value alignment yet. Add your own guardrails before exposing them to end users.
{% endhint %}


---

# 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/models/text-models.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.
