Instructions to use nace-ai/drex-dlm-Q8_0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use nace-ai/drex-dlm-Q8_0 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 nace-ai/drex-dlm-Q8_0:Q8_0
# Run inference directly in the terminal:
llama cli -hf nace-ai/drex-dlm-Q8_0:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf nace-ai/drex-dlm-Q8_0:Q8_0
# Run inference directly in the terminal:
llama cli -hf nace-ai/drex-dlm-Q8_0:Q8_0

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 nace-ai/drex-dlm-Q8_0:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf nace-ai/drex-dlm-Q8_0:Q8_0

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 nace-ai/drex-dlm-Q8_0:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf nace-ai/drex-dlm-Q8_0:Q8_0

Use Docker

docker model run hf.co/nace-ai/drex-dlm-Q8_0:Q8_0

LM Studio
Jan
Ollama
How to use nace-ai/drex-dlm-Q8_0 with Ollama:
```
ollama run hf.co/nace-ai/drex-dlm-Q8_0:Q8_0
```
Unsloth Desktop

How to use nace-ai/drex-dlm-Q8_0 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf nace-ai/drex-dlm-Q8_0:Q8_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": "nace-ai/drex-dlm-Q8_0:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use nace-ai/drex-dlm-Q8_0 with Docker Model Runner:
```
docker model run hf.co/nace-ai/drex-dlm-Q8_0:Q8_0
```

Lemonade

How to use nace-ai/drex-dlm-Q8_0 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull nace-ai/drex-dlm-Q8_0:Q8_0

Run and chat with the model

lemonade run user.drex-dlm-Q8_0-Q8_0

List all available models

lemonade list

Hermes Agent

How to use nace-ai/drex-dlm-Q8_0 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf nace-ai/drex-dlm-Q8_0:Q8_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 nace-ai/drex-dlm-Q8_0:Q8_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use nace-ai/drex-dlm-Q8_0 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf nace-ai/drex-dlm-Q8_0:Q8_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 "nace-ai/drex-dlm-Q8_0:Q8_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"

Drex DLM Q8_0 GGUF

This repository contains a Q8_0 conversion of Nace.AI's Drex DLM checkpoint. For the request schema, installation, native-server build, and runner configuration, see the source and serving README. The model uses the custom edlm architecture and POST /v1/systemone; a stock llama.cpp or Ollama installation is not a supported runner. Build the native edlm fork or the Ollama nace-edlm fork.

Artifact

File	Size	SHA-256
`drex-dlm-Q8_0.gguf`	8,713,715,232 bytes (8.115 GiB)	`04a592249958762d04ba0daf8e851eb89506e9fd0a38291e1e2ec9e6f6d61710`

The GGUF contains 254 Q8_0 tensors, 148 F32 tensors, and two F16 pointer-head matrices required by the decision scorer. It embeds general.license = cc-by-nc-4.0 and is 46.84% smaller on disk than the F16 GGUF.

Download the GGUF with hf download nace-ai/drex-dlm-Q8_0 drex-dlm-Q8_0.gguf --local-dir ./drex-dlm-q8, then point the native server at ./drex-dlm-q8/drex-dlm-Q8_0.gguf. On Apple Silicon, set GGML_METAL_TENSOR_DISABLE=1; use --embedding --pooling none -c 16384 -b 16384 -ub 16384 -np 1 --no-warmup for the tested native configuration. The recommended serving context is 16,384 tokens; the model's maximum position window is 32,768 tokens. Refer to the source README for the full command and request example.

Validation

Native serving was tested on an Apple M5 Pro with 48 GiB unified memory. The server loaded the Q8_0 GGUF at a 16,384-token context, completed requests, and returned the expected result for the short sample.

License

The model weights are released under CC BY-NC 4.0. The original code written by Nace.AI in this repository is under the MIT License.

Downloads last month: 52

GGUF

Model size

8B params

Architecture

edlm

Hardware compatibility

8-bit

Inference Providers NEW

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

Model tree for nace-ai/drex-dlm-Q8_0

Base model

nvidia/Efficient-DLM-8B

Finetuned

nace-ai/drex-dlm

Quantized

(1)

this model