Instructions to use TheDrummer/Rocinante-XL-16B-v1-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheDrummer/Rocinante-XL-16B-v1-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="TheDrummer/Rocinante-XL-16B-v1-GGUF",
	filename="Rocinante-XL-16B-v1a-Q2_K.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 TheDrummer/Rocinante-XL-16B-v1-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TheDrummer/Rocinante-XL-16B-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf TheDrummer/Rocinante-XL-16B-v1-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TheDrummer/Rocinante-XL-16B-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf TheDrummer/Rocinante-XL-16B-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/Rocinante-XL-16B-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf TheDrummer/Rocinante-XL-16B-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/Rocinante-XL-16B-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf TheDrummer/Rocinante-XL-16B-v1-GGUF:Q4_K_M

Use Docker

docker model run hf.co/TheDrummer/Rocinante-XL-16B-v1-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use TheDrummer/Rocinante-XL-16B-v1-GGUF with Ollama:
```
ollama run hf.co/TheDrummer/Rocinante-XL-16B-v1-GGUF:Q4_K_M
```

Unsloth Studio

How to use TheDrummer/Rocinante-XL-16B-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/Rocinante-XL-16B-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/Rocinante-XL-16B-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/Rocinante-XL-16B-v1-GGUF to start chatting

How to use TheDrummer/Rocinante-XL-16B-v1-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf TheDrummer/Rocinante-XL-16B-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/Rocinante-XL-16B-v1-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use TheDrummer/Rocinante-XL-16B-v1-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf TheDrummer/Rocinante-XL-16B-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/Rocinante-XL-16B-v1-GGUF:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use TheDrummer/Rocinante-XL-16B-v1-GGUF with Docker Model Runner:
```
docker model run hf.co/TheDrummer/Rocinante-XL-16B-v1-GGUF:Q4_K_M
```

Lemonade

How to use TheDrummer/Rocinante-XL-16B-v1-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull TheDrummer/Rocinante-XL-16B-v1-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Rocinante-XL-16B-v1-GGUF-Q4_K_M

List all available models

lemonade list

Broken LM STUDIO - Silly Tavern????

by uselessoldman - opened 8 days ago

Discussion

uselessoldman

8 days ago

I had been using this model in LM Studio as a server with Silly Tavern without any issues up until a few days ago when it started to corrupt when I edited or added data to the character card or past {{user}} prompts today it will not work at all - prompt template issue. I am using chat/ custom open ai-compatible in Silly Tavern and tried all the prompt templates in LM Studio inference without any success? This is not the only model that has started to misbehave over the past few days and originally I had thought it was a network issue but when I tested with Silly Tavern and LM Studio on the same machine (not over my network) its the same and some models corrupt some do not work at all. Anyone have similar experience or any advice and assistance since it has been driving me bonkers trying to fix? I have tried different note js? I do not really understand the core technical architecture behind AI inference just the basics and tried to get Ai to help me without any success - its as useless and lost as I am. I think it must be recent changes with LM Studio runtimes - I use Vulkan which has not been updated for some time?

artificialpossum

3 days ago

Just run this on koboldcpp instead of LM Studio and never look back.

uselessoldman

2 days ago

Would you believe Gemini (Google search) could not solve it but Copilot did? I had recently installed another GPU giving me 32Gb VRAM so I increased all the batch sizes which it appears Silly Tavern does not like you to do even if your GPU cards have 2Gb bandwidth and reduce concurrent connections to 1. Thankfully all is now back to normal and stable and my little AI workstation is a happy bunny again :)

uselessoldman changed discussion status to closed 2 days ago

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