Instructions to use badtheorylabs/BTL-3-Compact with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use badtheorylabs/BTL-3-Compact with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="badtheorylabs/BTL-3-Compact",
	filename="model/BTL-3-Compact-AVQ2.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use badtheorylabs/BTL-3-Compact 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 badtheorylabs/BTL-3-Compact
# Run inference directly in the terminal:
llama cli -hf badtheorylabs/BTL-3-Compact

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf badtheorylabs/BTL-3-Compact
# Run inference directly in the terminal:
llama cli -hf badtheorylabs/BTL-3-Compact

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 badtheorylabs/BTL-3-Compact
# Run inference directly in the terminal:
./llama-cli -hf badtheorylabs/BTL-3-Compact

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 badtheorylabs/BTL-3-Compact
# Run inference directly in the terminal:
./build/bin/llama-cli -hf badtheorylabs/BTL-3-Compact

Use Docker

docker model run hf.co/badtheorylabs/BTL-3-Compact

LM Studio
Jan

vLLM

How to use badtheorylabs/BTL-3-Compact with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "badtheorylabs/BTL-3-Compact"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "badtheorylabs/BTL-3-Compact",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/badtheorylabs/BTL-3-Compact

Ollama
How to use badtheorylabs/BTL-3-Compact with Ollama:
```
ollama run hf.co/badtheorylabs/BTL-3-Compact
```

Unsloth Studio

How to use badtheorylabs/BTL-3-Compact 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 badtheorylabs/BTL-3-Compact 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 badtheorylabs/BTL-3-Compact to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for badtheorylabs/BTL-3-Compact to start chatting

How to use badtheorylabs/BTL-3-Compact with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf badtheorylabs/BTL-3-Compact

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": "badtheorylabs/BTL-3-Compact"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use badtheorylabs/BTL-3-Compact with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf badtheorylabs/BTL-3-Compact

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 badtheorylabs/BTL-3-Compact

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use badtheorylabs/BTL-3-Compact with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf badtheorylabs/BTL-3-Compact

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 "badtheorylabs/BTL-3-Compact" \
  --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 badtheorylabs/BTL-3-Compact with Docker Model Runner:
```
docker model run hf.co/badtheorylabs/BTL-3-Compact
```

Lemonade

How to use badtheorylabs/BTL-3-Compact with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull badtheorylabs/BTL-3-Compact

Run and chat with the model

lemonade run user.BTL-3-Compact-{{QUANT_TAG}}

List all available models

lemonade list

BTL-3-Compact / docs /launch-btl3-compact.md

affableiq

Publish BTL-3 Compact package metadata

274ffba verified 8 days ago

preview code

Raw

History Blame Contribute Delete

5.23 kB

Launch BTL-3 Compact locally

This repository ships a relocatable macOS arm64 server bundle for the exact BTL-3 Compact AVQ2 artifact:

model: BTL-3-Compact-AVQ2.gguf
bytes: 8,392,369,600
SHA-256: 2ddf9527620a17a2a6739d184a7096c45712092e6589128792ec6254e94dc30c
runtime: llama.cpp build 9596, commit 9fcaed763

The model is external to the 28 MB runtime bundle. The launcher finds the model in artifacts/release, in the bundle's model directory, or at BTL3_MODEL.

Build and verify the bundle

Run this on an Apple Silicon Mac with Homebrew OpenSSL 3 installed:

rm -rf artifacts/runtime/BTL-3-Compact-macos-arm64
.venv/bin/python tools/build_macos_arm64_bundle.py

The builder copies the native dependency closure, rewrites absolute install names and rpaths, ad-hoc signs the result, includes dependency licenses, and writes bundle-manifest.json.

OpenAI-compatible API

Start the server:

BTL3_CTX_SIZE=4096 \
  artifacts/runtime/BTL-3-Compact-macos-arm64/bin/btl3-server

Check it:

curl -s http://127.0.0.1:8080/health
curl -s http://127.0.0.1:8080/v1/models

Call chat completions with any OpenAI-compatible client:

curl http://127.0.0.1:8080/v1/chat/completions \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "BTL-3",
    "messages": [{"role": "user", "content": "Write a retrying fetch helper."}],
    "stream": true
  }'

Useful configuration:

Variable	Default	Meaning
`BTL3_HOST`	`127.0.0.1`	Listen address
`BTL3_PORT`	`8080`	OpenAI API port
`BTL3_CTX_SIZE`	`32768`	Allocated context
`BTL3_PARALLEL`	`1`	Concurrent slots
`BTL3_GPU_LAYERS`	`99`	Layers requested on Metal
`BTL3_API_KEY`	unset	Optional local bearer key
`BTL3_MODEL_ALIASES`	`BTL-3`	Comma-separated transport aliases

On a 16 GB Mac, begin at 2K–4K context. A larger context raises KV-cache and working-memory requirements. The model's declared context length is not a promise that a particular device can allocate it.

LM Studio

LM Studio's official openai-compat-endpoint generator can target the local API. Its current model picker uses a fixed set of model IDs, so expose one of those IDs as a transport alias:

BTL3_API_KEY=btl3-local \
BTL3_MODEL_ALIASES='BTL-3,gpt-4.1-2025-04-14' \
BTL3_CTX_SIZE=4096 \
  artifacts/runtime/BTL-3-Compact-macos-arm64/bin/btl3-server

Then:

Install the official plugin from its LM Studio Hub page.
Set Override Base URL to http://127.0.0.1:8080/v1.
Set its API key to btl3-local.
Select gpt-4.1-2025-04-14.

That name is only protocol routing. The served model remains BTL-3; the integration does not claim it is an OpenAI model.

Ollama CLI

Stock Ollama does not load BTL-3's custom AVQ2 GGUF. Do not use ollama create. The supplied bridge lets the unmodified Ollama CLI speak to the native BTL-3 server through Ollama's local HTTP protocol.

In terminal one:

BTL3_CTX_SIZE=4096 \
  artifacts/runtime/BTL-3-Compact-macos-arm64/bin/btl3-server

In terminal two:

BTL3_CTX_SIZE=4096 \
  artifacts/runtime/BTL-3-Compact-macos-arm64/bin/btl3-ollama-bridge

Then use the installed Ollama CLI:

OLLAMA_HOST=http://127.0.0.1:11435 ollama list
OLLAMA_HOST=http://127.0.0.1:11435 ollama show btl3-compact:latest
OLLAMA_HOST=http://127.0.0.1:11435 ollama run btl3-compact:latest

The bridge implements /api/chat, /api/generate, /api/tags, /api/show, /api/ps, and /api/version. It maps streaming to Ollama NDJSON and preserves reasoning, tool calls, structured output, cancellation, token counts, and the common sampling options. Unsupported Ollama management endpoints return 501 instead of pretending to work.

Validation status

The packaged executable has passed native model load, graph reservation, /health, and /v1/models checks on Apple M2. Protocol tests pass using the real installed Ollama CLI and a deterministic OpenAI-compatible upstream.

The native Metal runtime accelerates AVQ2, affine INT4, the vocabulary head, and rescued and ordinary embedding rows without reconstructing dense weights. On a base Apple M2 with 16 GB unified memory, a clean full-model smoke measured 2.30 prompt tokens/second and 2.48 generated tokens/second at a 128-token allocated context. Treat those as one-device smoke results, not universal throughput claims. Context size, prompt length, thermal state, and Apple chip generation will change performance.

The exact GGUF has passed full-model native CUDA execution on an RTX PRO 6000 Blackwell Server Edition. Three repetitions measured 84.70 prompt tokens/second and 43.16 generated tokens/second with full GPU offload. This does not establish RTX 4090, RTX 5090, DGX Spark, or Windows compatibility; each target still requires its own packaged-runtime gate. Do not reuse the old Python/Triton H100 number as a native-runtime claim.

The native server may display roughly 7.7B n_params; that value counts packed stored elements, not the logical 27B architecture. Product metadata exposed by the bridge reports the logical model class.