Instructions to use leonardoeloi/mindreader-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use leonardoeloi/mindreader-models with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="leonardoeloi/mindreader-models",
	filename="Gemma-3-1B.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use leonardoeloi/mindreader-models 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 leonardoeloi/mindreader-models
# Run inference directly in the terminal:
llama cli -hf leonardoeloi/mindreader-models

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf leonardoeloi/mindreader-models
# Run inference directly in the terminal:
llama cli -hf leonardoeloi/mindreader-models

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 leonardoeloi/mindreader-models
# Run inference directly in the terminal:
./llama-cli -hf leonardoeloi/mindreader-models

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 leonardoeloi/mindreader-models
# Run inference directly in the terminal:
./build/bin/llama-cli -hf leonardoeloi/mindreader-models

Use Docker

docker model run hf.co/leonardoeloi/mindreader-models

LM Studio
Jan

vLLM

How to use leonardoeloi/mindreader-models with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "leonardoeloi/mindreader-models"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "leonardoeloi/mindreader-models",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/leonardoeloi/mindreader-models

Ollama
How to use leonardoeloi/mindreader-models with Ollama:
```
ollama run hf.co/leonardoeloi/mindreader-models
```

Unsloth Studio

How to use leonardoeloi/mindreader-models 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 leonardoeloi/mindreader-models 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 leonardoeloi/mindreader-models to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for leonardoeloi/mindreader-models to start chatting

Atomic Chat new
Docker Model Runner
How to use leonardoeloi/mindreader-models with Docker Model Runner:
```
docker model run hf.co/leonardoeloi/mindreader-models
```

Lemonade

How to use leonardoeloi/mindreader-models with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull leonardoeloi/mindreader-models

Run and chat with the model

lemonade run user.mindreader-models-{{QUANT_TAG}}

List all available models

lemonade list

Mind Reader — modelos locais (só-texto)

GGUFs só-texto usados pelo Mind Reader, app macOS de autocomplete inline local. São quantizações text-only (sem encoders de visão/áudio, sem mmproj → capabilities=['completion']), escolhidas por caberem residentes na RAM (o que dá latência baixa) — ver o achado de engenharia sobre model bloat multimodal.

Baixados direto pelo app via Ollama embutido: ollama pull hf.co/<este-repo>:<arquivo>.gguf.

Modelos

arquivo	base	params	quant	tamanho	papel
`Gemma-4-E4B.gguf`	Gemma 4 E4B	7.5B (efetivo 4B)	UD-Q5_K_XL	6.7 GB	qualidade máxima (16 GB+)
`Gemma-3-4B.gguf`	Gemma 3 4B	3.9B	Q4_K_M (imatrix)	2.5 GB	★ recomendado (4B real, enxuto)
`Gemma-4-E2B.gguf`	Gemma 4 E2B	4.6B (efetivo 2B)	Q4_K_M (imatrix)	3.5 GB	rápido
`Qwen-3-1.7B.gguf`	Qwen 3 1.7B	1.7B	Q4_K_M (imatrix)	1.1 GB	multilíngue minúsculo
`Gemma-3-1B.gguf`	Gemma 3 1B	1B	Q5_K_M (imatrix)	0.9 GB	mínimo (Macs de 8 GB)

Atribuição

Quantizações da comunidade (imatrix "i1" / unsloth "UD"): unsloth, bartowski, mradermacher. Modelos-base: Google (Gemma 3 / Gemma 4) e Alibaba (Qwen 3). Re-hospedados aqui para estabilidade da distribuição do Mind Reader; todos os direitos dos modelos-base permanecem dos autores.

Licenças (mapa por modelo em `LICENSE`)

Gemma 4 (E4B, E2B) + Qwen 3 → Apache License 2.0 (LICENSE-apache-2.0.txt).
Gemma 3 (4B, 1B) → Gemma Terms of Use (LICENSE-gemma-terms.txt); uso também sujeito à Gemma Prohibited Use Policy, repassada a jusante.

Ao baixar/usar um arquivo você concorda com a licença dele.

Ao baixar/usar estes arquivos você concorda com as licenças acima.

Downloads last month: 220

GGUF

Model size

1.0B params

Architecture

gemma3

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Mind Reader — modelos locais (só-texto)

Modelos

Atribuição

Licenças (mapa por modelo em LICENSE)

Licenças (mapa por modelo em `LICENSE`)