Instructions to use pavel-tolstyko/ggml-model-Q4_K_M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pavel-tolstyko/ggml-model-Q4_K_M with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="pavel-tolstyko/ggml-model-Q4_K_M",
	filename="ggml-model-Q4_K_M.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 pavel-tolstyko/ggml-model-Q4_K_M with llama.cpp:

Install from brew

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

Install from WinGet (Windows)

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

Use Docker

docker model run hf.co/pavel-tolstyko/ggml-model-Q4_K_M:Q4_K_M

LM Studio
Jan
Ollama
How to use pavel-tolstyko/ggml-model-Q4_K_M with Ollama:
```
ollama run hf.co/pavel-tolstyko/ggml-model-Q4_K_M:Q4_K_M
```

Unsloth Studio new

How to use pavel-tolstyko/ggml-model-Q4_K_M 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 pavel-tolstyko/ggml-model-Q4_K_M 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 pavel-tolstyko/ggml-model-Q4_K_M to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for pavel-tolstyko/ggml-model-Q4_K_M to start chatting

Docker Model Runner
How to use pavel-tolstyko/ggml-model-Q4_K_M with Docker Model Runner:
```
docker model run hf.co/pavel-tolstyko/ggml-model-Q4_K_M:Q4_K_M
```

Lemonade

How to use pavel-tolstyko/ggml-model-Q4_K_M with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull pavel-tolstyko/ggml-model-Q4_K_M:Q4_K_M

Run and chat with the model

lemonade run user.ggml-model-Q4_K_M-Q4_K_M

List all available models

lemonade list

Model Card for TinyLlama-1.1B-Chat-v1.0 (Quantized)

This is a quantized version of TinyLlama-1.1B-Chat-v1.0.

Performance Evaluation

The quantized model was tested on the hellaswag dataset with the following results:

Metric	Base Model	Quantized Model	Change
hellaswag accuracy	0.456	0.462	unchanged
hellaswag normalized accuracy	0.64	0.64	unchanged
eval time (GPU) - seconds	219.67	209.34	4.70% decrease

The quantized version of TinyLlama-1.1B-Chat-v1.0 maintains similar accuracy while achieving a 4.7% reduction in evaluation time. This evaluation was conducted using GPU resources on a subset of 100 hellaswag samples for expediency. For production purposes, it is recommended to perform a full evaluation.

Quantization Approach
The model was quantized to 4-bits using the Q4_K_M method with llama.cpp, specifically designed for optimized GPU performance. The following steps were used:

Convert the original model to GGUF format:

python ./llama.cpp/convert_hf_to_gguf.py ./llama.cpp/models/TinyLlama-1.1B-Chat-v1.0/

Quantize the GGUF model to 4-bit Q4_K_M:

./llama.cpp/build/bin/llama-quantize ./llama.cpp/models/TinyLlama-1.1B-Chat-v1.0/ggml-model-Q4_K_M.gguf q4_k_m

Downloads last month: 7

GGUF

Model size

1B params

Architecture

llama

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for pavel-tolstyko/ggml-model-Q4_K_M

Base model

TinyLlama/TinyLlama-1.1B-Chat-v1.0

Quantized

(145)

this model