Instructions to use adoslabs/liara-minerva-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use adoslabs/liara-minerva-7b 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 adoslabs/liara-minerva-7b:Q8_0
# Run inference directly in the terminal:
llama cli -hf adoslabs/liara-minerva-7b:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf adoslabs/liara-minerva-7b:Q8_0
# Run inference directly in the terminal:
llama cli -hf adoslabs/liara-minerva-7b:Q8_0

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 adoslabs/liara-minerva-7b:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf adoslabs/liara-minerva-7b:Q8_0

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 adoslabs/liara-minerva-7b:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf adoslabs/liara-minerva-7b:Q8_0

Use Docker

docker model run hf.co/adoslabs/liara-minerva-7b:Q8_0

LM Studio
Jan

vLLM

How to use adoslabs/liara-minerva-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "adoslabs/liara-minerva-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "adoslabs/liara-minerva-7b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/adoslabs/liara-minerva-7b:Q8_0

Ollama
How to use adoslabs/liara-minerva-7b with Ollama:
```
ollama run hf.co/adoslabs/liara-minerva-7b:Q8_0
```

Unsloth Studio

How to use adoslabs/liara-minerva-7b 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 adoslabs/liara-minerva-7b 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 adoslabs/liara-minerva-7b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for adoslabs/liara-minerva-7b to start chatting

Docker Model Runner
How to use adoslabs/liara-minerva-7b with Docker Model Runner:
```
docker model run hf.co/adoslabs/liara-minerva-7b:Q8_0
```

Lemonade

How to use adoslabs/liara-minerva-7b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull adoslabs/liara-minerva-7b:Q8_0

Run and chat with the model

lemonade run user.liara-minerva-7b-Q8_0

List all available models

lemonade list

Atomic Chat

adoslabs commited on Jul 17

Commit

e0985c1

verified ·

1 Parent(s): 1fb5615

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +39 -0

README.md ADDED Viewed

	@@ -0,0 +1,39 @@

+---
+license: apache-2.0
+base_model: sapienzanlp/Minerva-7B-instruct-v1.0
+language:
+  - it
+pipeline_tag: text-generation
+tags:
+  - liara
+  - gguf
+  - tool-calling
+  - italian
+  - minerva
+---
+# Liara Minerva-7B (GGUF Q8_0)
+**Liara Minerva 7B 🏛️ Italiano top (desktop)** — il modello più capace della fila
+on-device dell'app **Liara** (assistente personale locale e privata).
+Fine-tuning LoRA (SFT + KTO) di [Minerva-7B](https://huggingface.co/sapienzanlp/Minerva-7B-instruct-v1.0)
+(Sapienza NLP, italiano-first) sul dominio Liara: conversazione naturale, memoria
+personale e **tool-calling testuale ChatML** (`<tool_call>{"name": …, "arguments": …}</tool_call>`)
+sui 30 strumenti dell'app (email, agenda, note, meteo, calcoli, web, file, peer, telefono).
+- **Quantizzazione: Q8_0** (qualità piena vs FP16). Solo desktop (7B, RAM 12 GB+).
+- **Temperatura consigliata: 0.7** (conversazionale).
+- **Allenamento**: KTO dal checkpoint a eval-minimo (pre-overfit), 17 pacchetti di
+  rinforzi curati (anti-fabbricazione meteo/siti/date, identità, over-refusal,
+  over-tooling, contesto nei follow-up, previsioni→web_search) + biblioteca.
+- **Smoke a temperatura reale**: 10/10 casi.
+## Uso previsto
+GGUF per l'app Liara (zeli-local), eseguito con llama.cpp. Prompt ChatML, blocco
+`# Tools` nel system.
+## Licenza
+Apache-2.0 (come Minerva-7B di Sapienza NLP).