Instructions to use dikiyplayerpig/dpp-gpt-V2.1-flash-90m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dikiyplayerpig/dpp-gpt-V2.1-flash-90m with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="dikiyplayerpig/dpp-gpt-V2.1-flash-90m",
	filename="dpp-gptV2.1-flash-90m-f16.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 dikiyplayerpig/dpp-gpt-V2.1-flash-90m 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 dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf dikiyplayerpig/dpp-gpt-V2.1-flash-90m: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 dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf dikiyplayerpig/dpp-gpt-V2.1-flash-90m: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 dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M

Use Docker

docker model run hf.co/dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M

LM Studio
Jan

vLLM

How to use dikiyplayerpig/dpp-gpt-V2.1-flash-90m with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dikiyplayerpig/dpp-gpt-V2.1-flash-90m"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dikiyplayerpig/dpp-gpt-V2.1-flash-90m",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M

Ollama
How to use dikiyplayerpig/dpp-gpt-V2.1-flash-90m with Ollama:
```
ollama run hf.co/dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M
```

Unsloth Studio

How to use dikiyplayerpig/dpp-gpt-V2.1-flash-90m 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 dikiyplayerpig/dpp-gpt-V2.1-flash-90m 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 dikiyplayerpig/dpp-gpt-V2.1-flash-90m to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for dikiyplayerpig/dpp-gpt-V2.1-flash-90m to start chatting

Atomic Chat new
Docker Model Runner
How to use dikiyplayerpig/dpp-gpt-V2.1-flash-90m with Docker Model Runner:
```
docker model run hf.co/dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M
```

Lemonade

How to use dikiyplayerpig/dpp-gpt-V2.1-flash-90m with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull dikiyplayerpig/dpp-gpt-V2.1-flash-90m:Q4_K_M

Run and chat with the model

lemonade run user.dpp-gpt-V2.1-flash-90m-Q4_K_M

List all available models

lemonade list

⚡dpp-gpt V2.1 Flash (90m)

(🇺🇸 English / 🇷🇺 Русский)

This is the ultra-lightweight "Flash" version of the dpp-gpt-V2.1 series. It is a microscopic Large Language Model trained entirely from scratch.

⚙️ Model Details

Parameters: 93M
Layers / Hidden Size / Heads: 11 / 768 / 12
Context Length: 4096 tokens
Vocabulary Size: 32768
Format: GGUF / PyTorch (.pth)
License: Apache 2.0

📊 Training Data

Pre-training: ~~11.26 Billion tokens (~~121.1 tokens/parameter).
Fine-Tuning (SFT): >16.5M high-quality tokens generated primarily by Gemma 4 (26b/12b/4b), Qwen 3.5 (35b/4b), and complex code from DeepSeek v4 Flash.

🚀 Capabilities & Advantages

Text Processing & Math: Retains the ability to solve arithmetic operations step-by-step (addition/subtraction up to hundreds of thousands), though with lower overall accuracy compared to the 260M version due to its microscopic size.
Languages: Comprehension of Russian, English, and French.
Efficiency: Extremely fast inference on any CPU/GPU due to the 93M parameter count and GQA architecture.

💡 Prompting & System Prompt

The model uses a strict ChatML format.

Standard Mode (No thinking):

<|im_start|>user
[NOTHINK] {prompt}<|im_end|>
<|im_start|>assistant

Reasoning Mode ([THINK] token): To force the model to "think" and use logic before answering, modify the prompt template. If you are using LM Studio, simply type . right before your prompt (without a space).

<|im_start|>user
[THINK] {prompt}<|im_end|>
<|im_start|>assistant

🇷🇺 Описание на русском

Это сверхлегкая «Flash» версия в линейке dpp-gpt-V2.1. Микроскопическая большая языковая модель, обученная полностью с нуля.

⚙️ Детали модели

Параметры: 93M
Слои / Размерность / Головы: 11 / 768 / 12
Контекст: 4096 токенов
Размер словаря: 32768
Формат: GGUF / PyTorch (.pth)
Лицензия: Apache 2.0

📊 Обучающие данные

Претрейн (Pre-training): ~~11.26 млрд токенов (~~121.1 токена на параметр).
Файнтюнинг (SFT): >16.5 млн высококачественных токенов, сгенерированных в основном с помощью Gemma 4 (26b/12b/4b), немного Qwen 3.5 (35b/4b) и сложным кодом от DeepSeek v4 Flash.

🚀 Способности и преимущества

Обработка текста и математика: Сохраняет способность пошагово решать арифметические задачи (сложение/вычитание до сотен тысяч), однако из-за микроскопического размера общая точность ниже по сравнению с версией на 260M.
Языки: Понимание русского, английского и французского языков.
Эффективность: Чрезвычайно быстрый инференс на любом CPU/GPU благодаря размеру в 93М параметров и архитектуре GQA.

💡 Промпты и системный промпт

Модель строго использует формат ChatML.

Стандартный режим (Без размышлений):

<|im_start|>user
[NOTHINK] {prompt}<|im_end|>
<|im_start|>assistant

Режим размышления (токен [THINK]): Чтобы заставить модель "думать" и использовать логику перед ответом, измените шаблон промпта. Если вы используете LM Studio, просто напишите . прямо перед вашим запросом (без пробела).

<|im_start|>user
[THINK] {prompt}<|im_end|>
<|im_start|>assistant

Downloads last month: 211

GGUF

Model size

0.1B params

Architecture

llama

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit