Instructions to use teex-pt/AMALIA-9B-0626-DPO-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use teex-pt/AMALIA-9B-0626-DPO-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="teex-pt/AMALIA-9B-0626-DPO-GGUF",
	filename="AMALIA-9B-0626-DPO-Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use teex-pt/AMALIA-9B-0626-DPO-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M

Use Docker

docker model run hf.co/teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use teex-pt/AMALIA-9B-0626-DPO-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "teex-pt/AMALIA-9B-0626-DPO-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "teex-pt/AMALIA-9B-0626-DPO-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M

Ollama
How to use teex-pt/AMALIA-9B-0626-DPO-GGUF with Ollama:
```
ollama run hf.co/teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M
```

Unsloth Studio

How to use teex-pt/AMALIA-9B-0626-DPO-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for teex-pt/AMALIA-9B-0626-DPO-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for teex-pt/AMALIA-9B-0626-DPO-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for teex-pt/AMALIA-9B-0626-DPO-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use teex-pt/AMALIA-9B-0626-DPO-GGUF with Docker Model Runner:
```
docker model run hf.co/teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M
```

Lemonade

How to use teex-pt/AMALIA-9B-0626-DPO-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.AMALIA-9B-0626-DPO-GGUF-Q4_K_M

List all available models

lemonade list

AMALIA-9B-0626-DPO — GGUF

GGUF quantizations of amalia-llm/AMALIA-9B-0626-DPO, the Portuguese (pt-PT) 9B model presented on 1 July 2026. GGUF runs on any platform — Windows, Linux, macOS, NVIDIA/AMD GPUs or plain CPUs — via llama.cpp, Ollama, LM Studio, Jan, and anything built on them.

File	Quant	Size	Quality (measured)
`AMALIA-9B-0626-DPO-Q8_0.gguf`	Q8_0	9.1 GB	lossless in practice (−0.1% perplexity vs BF16, within noise)
`AMALIA-9B-0626-DPO-Q4_K_M.gguf`	Q4_K_M	5.2 GB	small loss (+2.7% perplexity vs BF16)

The chat template is embedded — chat frontends pick it up automatically. Prompt the model in European Portuguese; that is what it is tuned for.

Usage

Ollama — runs straight from this repo, no manual download:

ollama run hf.co/teex-pt/AMALIA-9B-0626-DPO-GGUF:Q4_K_M
# or, for maximum quality:
ollama run hf.co/teex-pt/AMALIA-9B-0626-DPO-GGUF:Q8_0

LM Studio: search for teex-pt/AMALIA-9B-0626-DPO-GGUF and pick a quant.

llama.cpp (pass --jinja so the embedded chat template, including the default system prompt, is fully honored):

llama-cli -m AMALIA-9B-0626-DPO-Q4_K_M.gguf --jinja

Default system prompt

When you don't supply a system message, these builds present the model with a factual identity (extended from the original model's one-liner): Amália, a 9B open-source (Apache 2.0) Portuguese model named after Amália Rodrigues, presented on 1 July 2026 — with an instruction to answer in the user's language (primarily pt-PT) and not to invent details about its origin. Without this, the base model confabulates its identity (we observed invented personas and funding programmes). It is applied via the GGUF-embedded chat template (llama.cpp --jinja, LM Studio) and via the repo's system file (Ollama). Supplying your own system message fully overrides it.

Findings: how quantization affects AMALIA-9B

We converted and benchmarked five variants of this model (GGUF and MLX, 4-bit and 8-bit) on an Apple M5 Pro (48 GB, llama.cpp b9850, mlx-lm 0.31.3). Perplexity for the GGUF family was measured with llama-perplexity over ~16.5k tokens of Portuguese Wikipedia prose, against the BF16 GGUF of the same model:

Variant	PPL	Δ vs BF16
BF16	7.779 ± 0.247	—
Q8_0 (this repo)	7.770 ± 0.247	−0.1% (noise)
Q4_K_M (this repo)	7.989 ± 0.251	+2.7%

Key findings:

Q8_0 is lossless in practice — its perplexity is statistically identical to full precision. If you have the memory (~10 GB), there is no quality reason to run anything less.
Q4_K_M is the best 4-bit option we measured — it degrades less (+2.7%) than a plain round-to-nearest 4-bit quantization of the same model (+4.3% in our parallel MLX experiments), thanks to the K-quant mixed-precision layout, at nearly the same size.
4-bit is cheap but not free — in greedy side-by-side tests, 4-bit variants showed occasional factual slips absent at 8-bit (e.g. misattributed literary works in a culture question). Fluency, grammar, translation, JSON structure and long-context retrieval (needle at ~2k tokens) were unaffected.
Speed tracks bytes — generation on an M5 Pro (Metal): Q4_K_M ≈ 48 tok/s, Q8_0 ≈ 30 tok/s, BF16 ≈ 16 tok/s; prompt processing ~1,050 tok/s (llama-bench, pp512/tg128). On NVIDIA GPUs expect substantially higher.

Rule of thumb: quality-critical work → Q8_0; interactive chat, drafting, or ≤8 GB machines → Q4_K_M.

Conversion pipeline, for reproduction: convert_hf_to_gguf.py (BF16) → llama-quantize (llama.cpp b9850).

Related repos

For native Apple Silicon (MLX) builds with full benchmark data and scripts:

teex-pt/AMALIA-9B-0626-DPO-MLX-8bit — near-lossless, 30 tok/s
teex-pt/AMALIA-9B-0626-DPO-MLX-4bit — fastest on Mac, 55+ tok/s
github.com/teex-pt/pt-amalia — benchmark scripts, conversion pipelines, findings, improvement plan

Attribution

All credit for the model goes to the AMALIA team — these repos are format conversions plus benchmark documentation. Original model: amalia-llm/AMALIA-9B-0626-DPO (Apache 2.0). EuroLLM-based, llama architecture, 32k context.

Downloads last month: 73

GGUF

Model size

9B params

Architecture

llama

Hardware compatibility

4-bit

8-bit

Model tree for teex-pt/AMALIA-9B-0626-DPO-GGUF

Base model

amalia-llm/AMALIA-9B-0626-SFT

Finetuned

amalia-llm/AMALIA-9B-0626-DPO

Quantized

(10)

this model

teex-pt
/

AMALIA-9B-0626-DPO-GGUF