Instructions to use Meridian-MRM/Novely-33m-chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Meridian-MRM/Novely-33m-chat 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 Meridian-MRM/Novely-33m-chat:Q5_K_M
# Run inference directly in the terminal:
llama cli -hf Meridian-MRM/Novely-33m-chat:Q5_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Meridian-MRM/Novely-33m-chat:Q5_K_M
# Run inference directly in the terminal:
llama cli -hf Meridian-MRM/Novely-33m-chat:Q5_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 Meridian-MRM/Novely-33m-chat:Q5_K_M
# Run inference directly in the terminal:
./llama-cli -hf Meridian-MRM/Novely-33m-chat:Q5_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 Meridian-MRM/Novely-33m-chat:Q5_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Meridian-MRM/Novely-33m-chat:Q5_K_M

Use Docker

docker model run hf.co/Meridian-MRM/Novely-33m-chat:Q5_K_M

Llama new
LM Studio
Jan

vLLM

How to use Meridian-MRM/Novely-33m-chat with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Meridian-MRM/Novely-33m-chat"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Meridian-MRM/Novely-33m-chat",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Meridian-MRM/Novely-33m-chat:Q5_K_M

Ollama
How to use Meridian-MRM/Novely-33m-chat with Ollama:
```
ollama run hf.co/Meridian-MRM/Novely-33m-chat:Q5_K_M
```
Unsloth Desktop
Docker Model Runner
How to use Meridian-MRM/Novely-33m-chat with Docker Model Runner:
```
docker model run hf.co/Meridian-MRM/Novely-33m-chat:Q5_K_M
```

Lemonade

How to use Meridian-MRM/Novely-33m-chat with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Meridian-MRM/Novely-33m-chat:Q5_K_M

Run and chat with the model

lemonade run user.Novely-33m-chat-Q5_K_M

List all available models

lemonade list

Atomic Chat

Novely-33M-Chat

A 32.5M parameter experimental language model trained from scratch and aligned via SFT.

Specs

Parameters: 32.5M
Layers: 6 | Dim: 512 | Heads: 8
Context Length: 512
Format: GGUF (FP16)

Limitations

Not a knowledge model. It does not know facts, dates, math, or code.
English only.
No safety tuning. Do not deploy it in any user-facing role.
Trained on public-domain literature — its default voice is Victorian/formal, not modern/casual.

Chat Template

Paste this into the model's chat template field for proper functionality.

{% for message in messages %}{% if message['role'] == 'system' %}### System:
{{ message['content'] }}

{% elif message['role'] == 'user' %}### Human:
{{ message['content'] }}

{% elif message['role'] == 'assistant' %}### Assistant:
{{ message['content'] }}

{% endif %}{% endfor %}### Assistant:

Recommended settings

Context: 512
Temperature: 0.7
Top-K: 35
Top-P: 0.90
Repetition Penalty: 1.2 (important)
Stop strings: <|endoftext|>, ### Human:

Downloads last month: 118

GGUF

Model size

32.5M params

Architecture

llama

Hardware compatibility

5-bit

8-bit

View +1 variant