Instructions to use VoxelSparrow/tamriel-actor-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use VoxelSparrow/tamriel-actor-7b 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 VoxelSparrow/tamriel-actor-7b:Q6_K
# Run inference directly in the terminal:
llama cli -hf VoxelSparrow/tamriel-actor-7b:Q6_K

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf VoxelSparrow/tamriel-actor-7b:Q6_K
# Run inference directly in the terminal:
llama cli -hf VoxelSparrow/tamriel-actor-7b:Q6_K

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 VoxelSparrow/tamriel-actor-7b:Q6_K
# Run inference directly in the terminal:
./llama-cli -hf VoxelSparrow/tamriel-actor-7b:Q6_K

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 VoxelSparrow/tamriel-actor-7b:Q6_K
# Run inference directly in the terminal:
./build/bin/llama-cli -hf VoxelSparrow/tamriel-actor-7b:Q6_K

Use Docker

docker model run hf.co/VoxelSparrow/tamriel-actor-7b:Q6_K

LM Studio
Jan

vLLM

How to use VoxelSparrow/tamriel-actor-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "VoxelSparrow/tamriel-actor-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "VoxelSparrow/tamriel-actor-7b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/VoxelSparrow/tamriel-actor-7b:Q6_K

Ollama
How to use VoxelSparrow/tamriel-actor-7b with Ollama:
```
ollama run hf.co/VoxelSparrow/tamriel-actor-7b:Q6_K
```

Unsloth Studio

How to use VoxelSparrow/tamriel-actor-7b 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 VoxelSparrow/tamriel-actor-7b 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 VoxelSparrow/tamriel-actor-7b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for VoxelSparrow/tamriel-actor-7b to start chatting

How to use VoxelSparrow/tamriel-actor-7b with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf VoxelSparrow/tamriel-actor-7b:Q6_K

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": "VoxelSparrow/tamriel-actor-7b:Q6_K"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use VoxelSparrow/tamriel-actor-7b with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf VoxelSparrow/tamriel-actor-7b:Q6_K

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 "VoxelSparrow/tamriel-actor-7b:Q6_K" \
  --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"

Docker Model Runner
How to use VoxelSparrow/tamriel-actor-7b with Docker Model Runner:
```
docker model run hf.co/VoxelSparrow/tamriel-actor-7b:Q6_K
```

Lemonade

How to use VoxelSparrow/tamriel-actor-7b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull VoxelSparrow/tamriel-actor-7b:Q6_K

Run and chat with the model

lemonade run user.tamriel-actor-7b-Q6_K

List all available models

lemonade list

Hermes Agent

How to use VoxelSparrow/tamriel-actor-7b with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf VoxelSparrow/tamriel-actor-7b:Q6_K

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 VoxelSparrow/tamriel-actor-7b:Q6_K

Run Hermes

hermes

Atomic Chat

tamriel-actor 7B (v1, GGUF)

A lore-tuned Elder Scrolls NPC model for Mantella, the Skyrim mod that gives NPCs an AI voice. It plays whichever NPC Mantella hands it, stays in character, and never leaves Tamriel: ask it something modern and it gets confused in character rather than stepping out of the fiction to explain.

You: What's your favourite phone?
Lydia: Phone? Is that some Dwemer trinket you found in a ruin? I carry a sword, my Thane, not curiosities.

What it is

Base: Qwen2.5-7B-Instruct (Apache-2.0), fine-tuned with QLoRA.
Trained on the series' own words: UESP, the Imperial Library, the ESO string dump and dialogue from the older games, grounded through Mantella's exact v0.14 prompt template so it behaves under the prompt it is actually deployed with.
This file: tamriel-actor-v1-Q6_K.gguf, a Q6_K quant that serves comfortably on a single 8GB+ GPU.

How to use it (llama.cpp)

llama-server -m tamriel-actor-v1-Q6_K.gguf --jinja -c 8192 -ngl 99 --flash-attn --host 0.0.0.0 --port 8080

Then in Mantella (v0.14): LLM Service = OpenAI Compatible, Service URL = http://<host>:8080/v1, model = any string, secret key = any non-empty string, max tokens 250, temperature 0.7. Leave the default Skyrim prompt (the model was trained on it). Keep Advanced Actions off.

No GPU? A one-click AWS CloudFormation stack that stands this up on your own account (you pay only AWS, no fee) is the intended companion; point its ModelUrl at this file.

Honest notes on quality

It is a good actor and a decent historian, not an oracle. It reliably stays in character and deflects anachronisms, and it knows a lot of lore, but at 7B it can be wrong on deep or obscure history and will occasionally invent something plausible. It was tuned for immersion first; treat its lore as a knowledgeable NPC's, not a wiki. (The build notes: fact-dense wiki/lore-book grounding beat training on authentic-but-vague in-game dialogue, which produced a lovely voice that forgot its facts.)

Licence and disclaimer

Licence: CC BY-NC 4.0. Free to use, share and build on for non-commercial purposes, with attribution. The Qwen2.5 base is Apache-2.0; this fine-tune and its data layer are non-commercial.

This is an unofficial fan project. It is not affiliated with, endorsed by, or connected to Bethesda Softworks or ZeniMax Media. The Elder Scrolls, Skyrim, Tamriel and all related names and lore are the property of ZeniMax/Bethesda. The model was trained on publicly available Elder Scrolls material for non-commercial fan use only. If you represent the rights holder and would like it taken down, open an issue on the repo.

Downloads last month: 70

GGUF

Model size

8B params

Architecture

qwen2

Hardware compatibility

6-bit

Model tree for VoxelSparrow/tamriel-actor-7b

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-7B-Instruct

Quantized

(386)

this model