Instructions to use Slaxikov/VokiLLM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Slaxikov/VokiLLM with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Slaxikov/VokiLLM",
	filename="imatrix-vokillm.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Slaxikov/VokiLLM with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Slaxikov/VokiLLM:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Slaxikov/VokiLLM:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Slaxikov/VokiLLM:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Slaxikov/VokiLLM: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 Slaxikov/VokiLLM:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Slaxikov/VokiLLM: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 Slaxikov/VokiLLM:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Slaxikov/VokiLLM:Q4_K_M

Use Docker

docker model run hf.co/Slaxikov/VokiLLM:Q4_K_M

LM Studio
Jan
Ollama
How to use Slaxikov/VokiLLM with Ollama:
```
ollama run hf.co/Slaxikov/VokiLLM:Q4_K_M
```

Unsloth Studio new

How to use Slaxikov/VokiLLM 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 Slaxikov/VokiLLM 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 Slaxikov/VokiLLM to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Slaxikov/VokiLLM to start chatting

Pi new

How to use Slaxikov/VokiLLM with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Slaxikov/VokiLLM:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Slaxikov/VokiLLM:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Slaxikov/VokiLLM with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Slaxikov/VokiLLM:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default Slaxikov/VokiLLM:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use Slaxikov/VokiLLM with Docker Model Runner:
```
docker model run hf.co/Slaxikov/VokiLLM:Q4_K_M
```

Lemonade

How to use Slaxikov/VokiLLM with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Slaxikov/VokiLLM:Q4_K_M

Run and chat with the model

lemonade run user.VokiLLM-Q4_K_M

List all available models

lemonade list

VokiLLM / README.md

Slaxikov

Update README.md

fc0f327 verified about 2 months ago

preview code

raw

history blame contribute delete

3.96 kB

	---
	license: apache-2.0
	language:
	- ru
	tags:
	- gguf
	- instruct
	---

	# VokiLLM-0.5B-Instruct (GGUF)

	Готовые файлы модели VokiLLM-0.5B-Instruct в формате GGUF: FP16 и квантовки для `llama.cpp`/LM Studio и других совместимых рантаймов.

	## Кратко

	- Архитектура: `qwen2`
	- Размер: ~0.5B (лейбл в GGUF: `630M`)
	- Контекст: 8192
	- Назначение: Instruct / чат

	## Файлы

	Файлы лежат в корне репозитория. В таблице ниже названия — это ссылки на скачивание.

	\| Файл \| Квантование \| Размер (bytes) \|
	\|---\|---:\|---:\|
	\| [`vokillm-0.5b-instruct-fp16.gguf`](vokillm-0.5b-instruct-fp16.gguf?download=true) \| FP16 \| 1266425856 \|
	\| [`vokillm-0.5b-instruct-q8_0.gguf`](vokillm-0.5b-instruct-q8_0.gguf?download=true) \| Q8_0 \| 675710976 \|
	\| [`vokillm-0.5b-instruct-q6_k.gguf`](vokillm-0.5b-instruct-q6_k.gguf?download=true) \| Q6_K \| 650379296 \|
	\| [`vokillm-0.5b-instruct-q5_k_m.gguf`](vokillm-0.5b-instruct-q5_k_m.gguf?download=true) \| Q5_K_M \| 522186752 \|
	\| [`vokillm-0.5b-instruct-q5_1.gguf`](vokillm-0.5b-instruct-q5_1.gguf?download=true) \| Q5_1 \| 521348096 \|
	\| [`vokillm-0.5b-instruct-q5_k_s.gguf`](vokillm-0.5b-instruct-q5_k_s.gguf?download=true) \| Q5_K_S \| 514810880 \|
	\| [`vokillm-0.5b-instruct-q5_0.gguf`](vokillm-0.5b-instruct-q5_0.gguf?download=true) \| Q5_0 \| 490475520 \|
	\| [`vokillm-0.5b-instruct-q4_k_m.gguf`](vokillm-0.5b-instruct-q4_k_m.gguf?download=true) \| Q4_K_M \| 491400192 \|
	\| [`vokillm-0.5b-instruct-q4_k_s.gguf`](vokillm-0.5b-instruct-q4_k_s.gguf?download=true) \| Q4_K_S \| 479064064 \|
	\| [`vokillm-0.5b-instruct-q4_1.gguf`](vokillm-0.5b-instruct-q4_1.gguf?download=true) \| Q4_1 \| 459602944 \|
	\| [`vokillm-0.5b-instruct-q3_k_l.gguf`](vokillm-0.5b-instruct-q3_k_l.gguf?download=true) \| Q3_K_L \| 445933568 \|
	\| [`vokillm-0.5b-instruct-iq4_nl.gguf`](vokillm-0.5b-instruct-iq4_nl.gguf?download=true) \| IQ4_NL \| 430880768 \|
	\| [`vokillm-0.5b-instruct-q4_0.gguf`](vokillm-0.5b-instruct-q4_0.gguf?download=true) \| Q4_0 \| 428730368 \|
	\| [`vokillm-0.5b-instruct-iq4_xs.gguf`](vokillm-0.5b-instruct-iq4_xs.gguf?download=true) \| IQ4_XS \| 428020736 \|
	\| [`vokillm-0.5b-instruct-q3_k_m.gguf`](vokillm-0.5b-instruct-q3_k_m.gguf?download=true) \| Q3_K_M \| 432041984 \|
	\| [`vokillm-0.5b-instruct-iq3_xxs.gguf`](vokillm-0.5b-instruct-iq3_xxs.gguf?download=true) \| IQ3_XXS \| 416919296 \|
	\| [`vokillm-0.5b-instruct-iq3_s.gguf`](vokillm-0.5b-instruct-iq3_s.gguf?download=true) \| IQ3_S \| 415182848 \|
	\| [`vokillm-0.5b-instruct-q2_k.gguf`](vokillm-0.5b-instruct-q2_k.gguf?download=true) \| Q2_K \| 415182848 \|
	\| [`vokillm-0.5b-instruct-q3_k_s.gguf`](vokillm-0.5b-instruct-q3_k_s.gguf?download=true) \| Q3_K_S \| 414838784 \|
	\| [`vokillm-0.5b-instruct-iq2_s.gguf`](vokillm-0.5b-instruct-iq2_s.gguf?download=true) \| IQ2_S \| 402312928 \|
	\| [`vokillm-0.5b-instruct-iq2_xs.gguf`](vokillm-0.5b-instruct-iq2_xs.gguf?download=true) \| IQ2_XS \| 400985056 \|
	\| [`vokillm-0.5b-instruct-iq2_xxs.gguf`](vokillm-0.5b-instruct-iq2_xxs.gguf?download=true) \| IQ2_XXS \| 398125024 \|
	\| [`vokillm-0.5b-instruct-iq1_s.gguf`](vokillm-0.5b-instruct-iq1_s.gguf?download=true) \| IQ1_S \| 392404960 \|

	## Как запустить

	### llama.cpp (CLI)

	Пример (Windows):

	```bash
	llama-cli.exe -m "vokillm-0.5b-instruct-q4_k_m.gguf" -p "Привет! Коротко объясни, что такое GGUF."
	```

	### LM Studio

	- Открой LM Studio → Models → Add model → выбери нужный `*.gguf`
	- Рекомендуемый стартовый вариант: `q4_k_m` (баланс скорость/качество)

	## Про квантовки IQ2/IQ1

	Квантовки `IQ2_` и `IQ1_S` сделаны с importance matrix (imatrix)* (это повышает качество для “экстремальных” квантовок).

	---
	license: apache-2.0
	language:
	- ru
	tags:
	- gguf
	- instruct
	---

	# VokiLLM-0.5B-Instruct (GGUF)

	Готовые файлы модели VokiLLM-0.5B-Instruct в формате GGUF: FP16 и квантовки для `llama.cpp`/LM Studio и других совместимых рантаймов.

	## Кратко

	- Архитектура: `qwen2`
	- Размер: ~0.5B (лейбл в GGUF: `630M`)
	- Контекст: 8192
	- Назначение: Instruct / чат

	## Файлы

	Файлы лежат в корне репозитория. В таблице ниже названия — это ссылки на скачивание.

	\| Файл \| Квантование \| Размер (bytes) \|
	\|---\|---:\|---:\|
	\| [`vokillm-0.5b-instruct-fp16.gguf`](vokillm-0.5b-instruct-fp16.gguf?download=true) \| FP16 \| 1266425856 \|
	\| [`vokillm-0.5b-instruct-q8_0.gguf`](vokillm-0.5b-instruct-q8_0.gguf?download=true) \| Q8_0 \| 675710976 \|
	\| [`vokillm-0.5b-instruct-q6_k.gguf`](vokillm-0.5b-instruct-q6_k.gguf?download=true) \| Q6_K \| 650379296 \|
	\| [`vokillm-0.5b-instruct-q5_k_m.gguf`](vokillm-0.5b-instruct-q5_k_m.gguf?download=true) \| Q5_K_M \| 522186752 \|
	\| [`vokillm-0.5b-instruct-q5_1.gguf`](vokillm-0.5b-instruct-q5_1.gguf?download=true) \| Q5_1 \| 521348096 \|
	\| [`vokillm-0.5b-instruct-q5_k_s.gguf`](vokillm-0.5b-instruct-q5_k_s.gguf?download=true) \| Q5_K_S \| 514810880 \|
	\| [`vokillm-0.5b-instruct-q5_0.gguf`](vokillm-0.5b-instruct-q5_0.gguf?download=true) \| Q5_0 \| 490475520 \|
	\| [`vokillm-0.5b-instruct-q4_k_m.gguf`](vokillm-0.5b-instruct-q4_k_m.gguf?download=true) \| Q4_K_M \| 491400192 \|
	\| [`vokillm-0.5b-instruct-q4_k_s.gguf`](vokillm-0.5b-instruct-q4_k_s.gguf?download=true) \| Q4_K_S \| 479064064 \|
	\| [`vokillm-0.5b-instruct-q4_1.gguf`](vokillm-0.5b-instruct-q4_1.gguf?download=true) \| Q4_1 \| 459602944 \|
	\| [`vokillm-0.5b-instruct-q3_k_l.gguf`](vokillm-0.5b-instruct-q3_k_l.gguf?download=true) \| Q3_K_L \| 445933568 \|
	\| [`vokillm-0.5b-instruct-iq4_nl.gguf`](vokillm-0.5b-instruct-iq4_nl.gguf?download=true) \| IQ4_NL \| 430880768 \|
	\| [`vokillm-0.5b-instruct-q4_0.gguf`](vokillm-0.5b-instruct-q4_0.gguf?download=true) \| Q4_0 \| 428730368 \|
	\| [`vokillm-0.5b-instruct-iq4_xs.gguf`](vokillm-0.5b-instruct-iq4_xs.gguf?download=true) \| IQ4_XS \| 428020736 \|
	\| [`vokillm-0.5b-instruct-q3_k_m.gguf`](vokillm-0.5b-instruct-q3_k_m.gguf?download=true) \| Q3_K_M \| 432041984 \|
	\| [`vokillm-0.5b-instruct-iq3_xxs.gguf`](vokillm-0.5b-instruct-iq3_xxs.gguf?download=true) \| IQ3_XXS \| 416919296 \|
	\| [`vokillm-0.5b-instruct-iq3_s.gguf`](vokillm-0.5b-instruct-iq3_s.gguf?download=true) \| IQ3_S \| 415182848 \|
	\| [`vokillm-0.5b-instruct-q2_k.gguf`](vokillm-0.5b-instruct-q2_k.gguf?download=true) \| Q2_K \| 415182848 \|
	\| [`vokillm-0.5b-instruct-q3_k_s.gguf`](vokillm-0.5b-instruct-q3_k_s.gguf?download=true) \| Q3_K_S \| 414838784 \|
	\| [`vokillm-0.5b-instruct-iq2_s.gguf`](vokillm-0.5b-instruct-iq2_s.gguf?download=true) \| IQ2_S \| 402312928 \|
	\| [`vokillm-0.5b-instruct-iq2_xs.gguf`](vokillm-0.5b-instruct-iq2_xs.gguf?download=true) \| IQ2_XS \| 400985056 \|
	\| [`vokillm-0.5b-instruct-iq2_xxs.gguf`](vokillm-0.5b-instruct-iq2_xxs.gguf?download=true) \| IQ2_XXS \| 398125024 \|
	\| [`vokillm-0.5b-instruct-iq1_s.gguf`](vokillm-0.5b-instruct-iq1_s.gguf?download=true) \| IQ1_S \| 392404960 \|

	## Как запустить

	### llama.cpp (CLI)

	Пример (Windows):

	```bash
	llama-cli.exe -m "vokillm-0.5b-instruct-q4_k_m.gguf" -p "Привет! Коротко объясни, что такое GGUF."
	```

	### LM Studio

	- Открой LM Studio → Models → Add model → выбери нужный `*.gguf`
	- Рекомендуемый стартовый вариант: `q4_k_m` (баланс скорость/качество)

	## Про квантовки IQ2/IQ1

	Квантовки `IQ2_` и `IQ1_S` сделаны с importance matrix (imatrix)* (это повышает качество для “экстремальных” квантовок).