Instructions to use RegalInvader/Marco-Nano-Instruct-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use RegalInvader/Marco-Nano-Instruct-GGUF 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 RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0
# Run inference directly in the terminal:
llama cli -hf RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0
# Run inference directly in the terminal:
llama cli -hf RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0
# Run inference directly in the terminal:
./build/bin/llama cli -hf RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

Use Docker

docker model run hf.co/RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

Llama new
LM Studio
Jan

vLLM

How to use RegalInvader/Marco-Nano-Instruct-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RegalInvader/Marco-Nano-Instruct-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RegalInvader/Marco-Nano-Instruct-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

Ollama
How to use RegalInvader/Marco-Nano-Instruct-GGUF with Ollama:
```
ollama run hf.co/RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0
```
Unsloth Desktop

How to use RegalInvader/Marco-Nano-Instruct-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use RegalInvader/Marco-Nano-Instruct-GGUF with Docker Model Runner:
```
docker model run hf.co/RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0
```

Lemonade

How to use RegalInvader/Marco-Nano-Instruct-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

Run and chat with the model

lemonade run user.Marco-Nano-Instruct-GGUF-TQ1_0

List all available models

lemonade list

Hermes Agent

How to use RegalInvader/Marco-Nano-Instruct-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

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 RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use RegalInvader/Marco-Nano-Instruct-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "RegalInvader/Marco-Nano-Instruct-GGUF:TQ1_0" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Marco-Nano-Instruct-GGUF / code /tests /test_gptq.py

RegalInvader

Release: Marco-Nano-Instruct 1-, 2- and 3-bit GGUF for phones

e5235fb verified 5 days ago

Raw History Blame Contribute Delete

1.72 kB

"""GPTQ-style ternary latents (T5.0) on random correlated inputs."""
import torch

from lowbit.quant.gptq import gptq_ternary
from lowbit.quant.ternary import QuantSpec, fake_quant

SPEC = QuantSpec(method="absmean", group_size=128)


def data(E=3, R=32, C=256, n=2048, seed=0):
    g = torch.Generator().manual_seed(seed)
    w = torch.randn(E, R, C, generator=g) * 0.05
    mix = torch.randn(C, C, generator=g) / C ** 0.5 + torch.eye(C)  # correlated inputs, as real activations are
    x = torch.randn(E, n, C, generator=g) @ mix
    return w, x, x.transpose(1, 2) @ x


def out_err(w, w_hat, x):
    return float(((x @ (w - w_hat).transpose(1, 2)) ** 2).sum())


def test_gptq_beats_round_to_nearest_on_output_error():
    w, x, h = data()
    latent, stats = gptq_ternary(w, h, SPEC)
    rtn = out_err(w, fake_quant(w, SPEC), x)
    gq = out_err(w, fake_quant(latent, SPEC), x)
    assert gq < 0.8 * rtn, (gq, rtn)
    assert stats["code_mismatch"] < 0.01  # the trainer's quantiser re-derives (almost) GPTQ's codes from the latent


def test_identity_hessian_is_round_to_nearest():
    w, _, _ = data()
    h = torch.eye(w.shape[-1]).expand(w.shape[0], -1, -1).clone()
    latent, _ = gptq_ternary(w, h, SPEC, damp=0.0)
    assert torch.equal(fake_quant(latent, SPEC), fake_quant(w, SPEC))  # no cross-column terms: nothing to compensate


def test_ill_conditioned_hessian_does_not_fail():
    w, x, h = data(n=64)  # fewer samples than inputs: rank-deficient
    x[:, :, 3] *= 1e4     # one massive activation dimension, as real hidden states have
    h = x.transpose(1, 2) @ x
    latent, stats = gptq_ternary(w, h, SPEC)
    assert torch.isfinite(latent).all() and stats["identity_fallback"] <= w.shape[0]

	"""GPTQ-style ternary latents (T5.0) on random correlated inputs."""
	import torch

	from lowbit.quant.gptq import gptq_ternary
	from lowbit.quant.ternary import QuantSpec, fake_quant

	SPEC = QuantSpec(method="absmean", group_size=128)


	def data(E=3, R=32, C=256, n=2048, seed=0):
	g = torch.Generator().manual_seed(seed)
	w = torch.randn(E, R, C, generator=g) * 0.05
	mix = torch.randn(C, C, generator=g) / C ** 0.5 + torch.eye(C) # correlated inputs, as real activations are
	x = torch.randn(E, n, C, generator=g) @ mix
	return w, x, x.transpose(1, 2) @ x


	def out_err(w, w_hat, x):
	return float(((x @ (w - w_hat).transpose(1, 2)) ** 2).sum())


	def test_gptq_beats_round_to_nearest_on_output_error():
	w, x, h = data()
	latent, stats = gptq_ternary(w, h, SPEC)
	rtn = out_err(w, fake_quant(w, SPEC), x)
	gq = out_err(w, fake_quant(latent, SPEC), x)
	assert gq < 0.8 * rtn, (gq, rtn)
	assert stats["code_mismatch"] < 0.01 # the trainer's quantiser re-derives (almost) GPTQ's codes from the latent


	def test_identity_hessian_is_round_to_nearest():
	w, _, _ = data()
	h = torch.eye(w.shape[-1]).expand(w.shape[0], -1, -1).clone()
	latent, _ = gptq_ternary(w, h, SPEC, damp=0.0)
	assert torch.equal(fake_quant(latent, SPEC), fake_quant(w, SPEC)) # no cross-column terms: nothing to compensate


	def test_ill_conditioned_hessian_does_not_fail():
	w, x, h = data(n=64) # fewer samples than inputs: rank-deficient
	x[:, :, 3] *= 1e4 # one massive activation dimension, as real hidden states have
	h = x.transpose(1, 2) @ x
	latent, stats = gptq_ternary(w, h, SPEC)
	assert torch.isfinite(latent).all() and stats["identity_fallback"] <= w.shape[0]