Instructions to use TheDrummer/Artemis-31B-v1-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 TheDrummer/Artemis-31B-v1-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 TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf TheDrummer/Artemis-31B-v1-GGUF: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 TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf TheDrummer/Artemis-31B-v1-GGUF: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 TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M

Use Docker

docker model run hf.co/TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use TheDrummer/Artemis-31B-v1-GGUF with Ollama:
```
ollama run hf.co/TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M
```

Unsloth Studio

How to use TheDrummer/Artemis-31B-v1-GGUF 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 TheDrummer/Artemis-31B-v1-GGUF 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 TheDrummer/Artemis-31B-v1-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for TheDrummer/Artemis-31B-v1-GGUF to start chatting

How to use TheDrummer/Artemis-31B-v1-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use TheDrummer/Artemis-31B-v1-GGUF with Docker Model Runner:
```
docker model run hf.co/TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M
```

Lemonade

How to use TheDrummer/Artemis-31B-v1-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Artemis-31B-v1-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use TheDrummer/Artemis-31B-v1-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 TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M

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 TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use TheDrummer/Artemis-31B-v1-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M

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 "TheDrummer/Artemis-31B-v1-GGUF:Q4_K_M" \
  --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"

Stop the presses! 🛑

by McG-221 - opened 23 days ago

Discussion

McG-221

23 days ago

This... this is it! Consign the others to the flames and: RELEASE. THIS. CRACKER!

TheDrummer

Owner 23 days ago

You don't like the newer releases?

Xominous

23 days ago

L has been the best in my testing/benchmarking. Excellent work!

McG-221

22 days ago

•

edited 22 days ago

You don't like the newer releases?

L has been the best in my testing/benchmarking. Excellent work!

Hey everybody, so I did an in-depth A/B test between v1h and v1l, both tested in Q8_0. Imho, v1h surpasses the first-party Google release in quality, both in language diversity (no corporate boilerplate speech), as well as in nuance, all while maintaining reasoning capabilities (which broke for me on v1n, for example; it still reasons, but the results are wrong, that's what I meant to say).
v1l: shorter, less nuanced answers, missing text layouts (headers, attributes like star ratings or ratings in general, text dividers).

Use cases vary from person to person, but I firmly continue to stand with v1h ✌️🤠🚀

FluffyKaeloky

20 days ago

•

edited 20 days ago

I've tested this model in Q5. Llama.cpp backend.

I must say I can't see what's going on with it. This is definitely not 'it' for me. The prose is good, sure, but from my tests in thinking mode, I've had many failures to end the thinking tag correctly, and many refusals over 'safe_rules'.
Even when the model generated correctly, it had the same profile I could find on other Gemma 4 fine-tunes, namely a gain in vocabulary and prose compared to the abliterated/uncensored base but a loss of instruction and detail tracking.

So far, I've tested many fine-tunes, and I have yet to find something that comes close in terms of balance between prose and attention to sophosympatheia/Glistening-Gem-31B-v1.0. I don't know what's going on in that merge but I would recommend taking a look at it.
Ironically, it seems to have some this model in the merge as well.

This is of course just my own experience with it, and my personal opinion.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment