Instructions to use Murban35/Macron_T2_32K with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Murban35/Macron_T2_32K with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Murban35/Macron_T2_32K",
	filename="Macron_T2_32K-i2s-f16.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Murban35/Macron_T2_32K with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Murban35/Macron_T2_32K:F16
# Run inference directly in the terminal:
llama-cli -hf Murban35/Macron_T2_32K:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Murban35/Macron_T2_32K:F16
# Run inference directly in the terminal:
llama-cli -hf Murban35/Macron_T2_32K:F16

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 Murban35/Macron_T2_32K:F16
# Run inference directly in the terminal:
./llama-cli -hf Murban35/Macron_T2_32K:F16

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 Murban35/Macron_T2_32K:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Murban35/Macron_T2_32K:F16

Use Docker

docker model run hf.co/Murban35/Macron_T2_32K:F16

LM Studio
Jan
Ollama
How to use Murban35/Macron_T2_32K with Ollama:
```
ollama run hf.co/Murban35/Macron_T2_32K:F16
```

Unsloth Studio

How to use Murban35/Macron_T2_32K 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 Murban35/Macron_T2_32K 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 Murban35/Macron_T2_32K to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Murban35/Macron_T2_32K to start chatting

Atomic Chat new
Docker Model Runner
How to use Murban35/Macron_T2_32K with Docker Model Runner:
```
docker model run hf.co/Murban35/Macron_T2_32K:F16
```

Lemonade

How to use Murban35/Macron_T2_32K with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Murban35/Macron_T2_32K:F16

Run and chat with the model

lemonade run user.Macron_T2_32K-F16

List all available models

lemonade list

Macron T2 32K

Mobile GGUF release for the Galvaniq Android app.

Files

Macron_T2_32K-i2s-f16.gguf
Macron_T2_32K-i2s-f16-meta.json

Build

Runtime: bitnet.cpp
Format: I2_S ternary linear tensors with F16 token embedding
Context length: 32768
RoPE: YaRN, factor 8.0, original context 4096, theta 500000
Size: 1187801440 bytes
SHA256: e7901384bdfc226018c430fa9e5ea88051a682d06c02e33f67cb974f1a046b66

Verification

Tensor gate: 210 I2_S tensors, 0 Q8_0 tensors
token_embd.weight: F16
output.weight: absent
GGUF metadata includes bitnet-25.context_length = 32768
Modal smoke test loaded successfully with n_ctx = 32768

This release is intended for the Galvaniq app's BitNet runtime path, not as a generic llama.cpp compatibility claim.

Downloads last month: 17

GGUF

Model size

2B params

Architecture

bitnet-25

Hardware compatibility

16-bit

Inference Providers NEW

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