aac board generator 770m ptbr needs about 1.3 GB of memory at Q4_K_M (a 0.6 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 5 quants, from 0.6 GB (Q4_K_M) to 1.5 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 0.9 GB.
Compact Brazilian‑Portuguese AAC board generator and editor
aac board generator 770m ptbr is a 771.6M‑parameter Gemma‑3 model distilled for Augmentative and Alternative Communication. It drafts pictogram‑board word lists and follows in‑assistant editing instructions. The model runs on CPU and is available as GGUF builds for llama.cpp.
Good at
- Board generation from activity themes
- Instruction‑following for board edits
- Fast CPU inference (~4 s per board on 4 cores)
Limits
- No GPU support
- Limited to 32768‑token context
- No vision capabilities
Best for
- Creating AAC boards in pt‑BR
- Editing existing AAC boards
- Integrating with Papuguinho workflow
Who it’s for. Developers and designers building AAC tools in Brazilian Portuguese. The model is suited for applications that need quick, deterministic board drafts and edits without GPU resources.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 0.6 GB | 1.3 GB | Fits | Get file → | |
| Q5_K_M | 0.6 GB | 1.3 GB | Fits | Get file → | |
| Q6_K | 0.8 GB | 1.5 GB | Fits | Get file → | |
| Q8_0 | 0.8 GB | 1.5 GB | Fits | Get file → | |
| F16 | 1.5 GB | 2.3 GB | Fits | Get file → |
For 8 GB VRAM use Q4_K_M or Q5_K_M (≈1.3 GB RAM). For 16 GB VRAM Q6_K or Q8_0 (≈1.5 GB RAM). The F16 build needs ≈2.3 GB RAM and is best for higher precision when memory allows.
What GPU do you need for aac board generator 770m ptbr?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is F16 (1.5 GB file, about 2.3 GB in use).
Which aac board generator 770m ptbr quant should you choose?
Q4_K_M is the usual starting point for aac board generator 770m ptbr: it balances file size and quality well. Here it is a 0.6 GB file.
The Q8_0 file is 0.8 GB, about 1.5 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about aac board generator 770m ptbr
How much VRAM does aac board generator 770m ptbr need?
About 1.3 GB at Q4_K_M with an 8K context window. That is the 0.6 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run aac board generator 770m ptbr on an 8 GB GPU?
Yes, with F16: the 1.5 GB file needs about 2.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of aac board generator 770m ptbr GGUF?
It depends on the quant: Q4_K_M is 0.6 GB, Q5_K_M is 0.6 GB, Q6_K is 0.8 GB, Q8_0 is 0.8 GB, F16 is 1.5 GB.
What context length does aac board generator 770m ptbr support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.9 GB at 32K.
Where can I download aac board generator 770m ptbr GGUF?
The files are on Hugging Face in the tardellirs/aac-board-generator-770m-ptbr-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does aac board generator 770m ptbr use?
Hugging Face lists the license as gemma. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run this model on 8 GB VRAM?
Yes, the Q4_K_M or Q5_K_M quantizations use about 1.3 GB RAM at 8 k context, fitting comfortably in 8 GB VRAM.
Does the model support vision tasks?
No, the model is text‑only and has no vision capabilities.
What license applies to this model?
The model is released under the Gemma license, inherited from Gemma 3.