---
language:
- pt
license: gemma
library_name: gguf
pipeline_tag: text-generation
base_model: tardellirs/aac-board-generator-770m-ptbr
tags:
- gguf
- llama-cpp
- text-generation
- gemma3
- aac
- augmentative-alternative-communication
- arasaac
- board-generation
- brazilian-portuguese
- cpu
---

# AAC Board Generator 770M (pt-BR) — GGUF

![Runs on](https://img.shields.io/badge/runs%20on-CPU-success)
[![Model](https://img.shields.io/badge/model-card-blue)](https://huggingface.co/tardellirs/aac-board-generator-770m-ptbr)
[![License](https://img.shields.io/badge/license-Gemma-lightgrey)](https://ai.google.dev/gemma/terms)
[![Used in Papuguinho](https://img.shields.io/badge/used%20in-Papuguinho-ff69b4)](https://www.papuguinho.com)

`llama.cpp` **GGUF builds** of [**aac-board-generator-770m-ptbr**](https://huggingface.co/tardellirs/aac-board-generator-770m-ptbr)
— a 771.6M Brazilian-Portuguese **multi-function AAC** model that drafts **pictogram-board** word lists *and* follows
in-assistant board-editing instructions (add / remove / generate N), on **CPU**. See the
[**model card**](https://huggingface.co/tardellirs/aac-board-generator-770m-ptbr) for what it does, training, and
evaluation.

## Files

| File | Quant | Size | Notes |
|---|---|---|---|
| `aac-board-generator-770m-ptbr-Q4_K_M.gguf` | Q4_K_M | 561 MB | smallest; slight quality drop |
| `aac-board-generator-770m-ptbr-Q5_K_M.gguf` | Q5_K_M | 606 MB | good size/quality balance |
| `aac-board-generator-770m-ptbr-Q6_K.gguf`   | Q6_K   | 766 MB | near-lossless |
| `aac-board-generator-770m-ptbr-Q8_0.gguf`   | Q8_0   | 824 MB | **recommended** — safe near-lossless |
| `aac-board-generator-770m-ptbr-f16.gguf`    | f16    | 1.55 GB | full precision reference |

At **Q8_0** the model needs on the order of **~0.9–1.1 GB RAM** to serve (weights + a small context) and produces a
~12-item board in **about 4 s on 4 CPU threads**.

## Usage

```bash
llama-server -m aac-board-generator-770m-ptbr-Q8_0.gguf -t 4 -c 2048 --host 127.0.0.1 --port 8080
```

Gemma has no system role — put the instruction in the user turn and call `/completion`:

```python
import requests
INSTR = ("Você monta pranchas de CAA (pictogramas, pt-BR). Para o PEDIDO, liste ~12 itens concretos e relevantes, "
         "um por linha, no formato palavra|tipo|sinônimos (tipo: v/s/a/e/l/p). Só a lista.")
pedido = "monta uma prancha de brincar no parquinho"
prompt = f"<start_of_turn>user\n{INSTR}\n\nPEDIDO: {pedido}<end_of_turn>\n<start_of_turn>model\n"
print(requests.post("http://127.0.0.1:8080/completion",
                    json={"prompt": prompt, "temperature": 0, "n_predict": 320}).json()["content"])
```

Generation is **greedy / deterministic** (`temperature 0`). Output is one item per line, `word|type|synonyms`.

## Notes on the build (why these GGUFs are faithful)

This model is a **vocabulary-trimmed** Gemma 3 (262k → 64k tokens). A reconstructed SentencePiece tokenizer is *not*
bit-exact for a trimmed vocabulary and would silently degrade greedy generation. These GGUFs instead use the
**BPE `tokenizer.json`** path, which tokenizes **identically** to the original — so the quantized model reproduces the
untrimmed parent's outputs. Verified: identical token counts and identical boards vs. the fp16 reference.

## License & attribution

Released under the **[Gemma license](https://ai.google.dev/gemma/terms)** (inherited from Gemma 3). Built for the
**ARASAAC** ecosystem (CC BY-NC-SA content). See the
[model card](https://huggingface.co/tardellirs/aac-board-generator-770m-ptbr) for full terms.

Developed for and used in **Papuguinho** — [www.papuguinho.com](https://www.papuguinho.com).