Instructions to use FirstPrinciplesLabs/quark_quantized_weight_only-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 FirstPrinciplesLabs/quark_quantized_weight_only-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 FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0
# Run inference directly in the terminal:
llama cli -hf FirstPrinciplesLabs/quark_quantized_weight_only-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 FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0
# Run inference directly in the terminal:
llama cli -hf FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0
# Run inference directly in the terminal:
./build/bin/llama cli -hf FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0

Use Docker

docker model run hf.co/FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0

Llama new
LM Studio
Jan

vLLM

How to use FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0

Ollama
How to use FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0 with Ollama:
```
ollama run hf.co/FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0
```
Unsloth Desktop

How to use FirstPrinciplesLabs/quark_quantized_weight_only-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 FirstPrinciplesLabs/quark_quantized_weight_only-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": "FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0 with Docker Model Runner:
```
docker model run hf.co/FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0
```

Lemonade

How to use FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0

Run and chat with the model

lemonade run user.quark_quantized_weight_only-Q8_0-Q8_0

List all available models

lemonade list

Hermes Agent

How to use FirstPrinciplesLabs/quark_quantized_weight_only-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 FirstPrinciplesLabs/quark_quantized_weight_only-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 FirstPrinciplesLabs/quark_quantized_weight_only-Q8_0:Q8_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use FirstPrinciplesLabs/quark_quantized_weight_only-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 FirstPrinciplesLabs/quark_quantized_weight_only-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 "FirstPrinciplesLabs/quark_quantized_weight_only-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"

Matthew Tchouikine commited on 5 days ago

Commit

fb64230

1 Parent(s): a8c24fc

cleanup

Browse files

Files changed (1) hide show

README.md +25 -30

README.md CHANGED Viewed

@@ -3,7 +3,7 @@ license: apache-2.0
 base_model:
 - FirstPrinciplesLabs/quark
 base_model_relation: quantized
-pipeline_tag: image-text-to-text
 language:
 - en
 tags:
@@ -15,52 +15,47 @@ tags:
 - math
 ---
-# Quark GGUF
-8-bit GGUF of [Quark](https://huggingface.co/FirstPrinciplesLabs/quark), a 9B reasoning model for physics,
-science and maths by [FirstPrinciples](https://firstprinciples.com/), for
 [llama.cpp](https://github.com/ggml-org/llama.cpp) and tools built on it (Ollama, LM Studio).
-| File | Contents | Quantization | Size |
 |---|---|---|---|
-| `quark-Q8_0.gguf` | language model | Q8_0 (8-bit) | 9.8 GB |
-| `mmproj-quark-BF16.gguf` | vision encoder, for image input | BF16 (16-bit) | 0.9 GB |
-## Results and speed
-The Q8_0 GGUF matches the original model on GPQA Diamond, AIME 2025, MMLU-Pro Physics and Humanity's Last
-Exam. On the Quark model card: [benchmark scores](https://huggingface.co/FirstPrinciplesLabs/quark#results)
-and [speed with llama.cpp and vLLM](https://huggingface.co/FirstPrinciplesLabs/quark#speed) (Q8_0 vs bf16,
-on an NVIDIA H200 and a MacBook Pro).
 ## Usage
 ```bash
-# Full thinking budget (needs about 15 GB of GPU memory):
 llama-server -hf FirstPrinciplesLabs/quark-GGUF:Q8_0 --jinja -ngl 99 -c 135168
-# 16 GB Macs: a 32K context (answers longer than that are cut off)
 llama-server -hf FirstPrinciplesLabs/quark-GGUF:Q8_0 --jinja -ngl 99 -c 32768
 ```
-`--jinja` uses the chat template stored in the file, which handles thinking mode. `-ngl 99` puts all
-layers on the GPU. `-c` is the context size, prompt and answer together: 135,168 tokens leaves room for a
-131,072-token answer plus the prompt.
-We recommend an output budget of up to 131,072 tokens and the sampling settings from the
-[Quark model card](https://huggingface.co/FirstPrinciplesLabs/quark#usage)
 (thinking mode: temperature 1.0, top_p 0.95, top_k 20, presence penalty 1.5).
-`-hf` also downloads the vision encoder, so the server accepts images (as `image_url` parts of a chat
-message, or attached in its web page). Add `--no-mmproj` for text only. With local files:
-```bash
-llama-server -m quark-Q8_0.gguf --mmproj mmproj-quark-BF16.gguf --jinja -ngl 99 -c 32768
-```
-Image input uses Qwen3.5-9B's vision encoder, which Quark's fine-tuning did not change. We checked it on
-simple images (text, labelled diagrams, charts) but did not benchmark it.
 ## License
 Apache 2.0

 base_model:
 - FirstPrinciplesLabs/quark
 base_model_relation: quantized
+pipeline_tag: text-generation
 language:
 - en
 tags:
 - math
 ---
+# Quark int8
+[Quark](https://huggingface.co/FirstPrinciplesLabs/quark), a 9B reasoning model for physics, science and
+maths by [FirstPrinciples](https://firstprinciples.com/), quantized to int8 for
 [llama.cpp](https://github.com/ggml-org/llama.cpp) and tools built on it (Ollama, LM Studio).
+| File | Format | Size | Input |
 |---|---|---|---|
+| `quark-Q8_0.gguf` | Q8_0 (int8) | 9.8 GB | text |
+Q8_0 is llama.cpp's int8 format: weights are stored as 8-bit integers, with one scale shared by each
+block of 32.
+## Results
+The int8 version matches the original bf16 model on GPQA Diamond, AIME 2025, MMLU-Pro Physics and
+Humanity's Last Exam, and generates faster. See the Quark model card for the
+[benchmark scores](https://huggingface.co/FirstPrinciplesLabs/quark#results) and
+[throughput](https://huggingface.co/FirstPrinciplesLabs/quark#throughput) on an NVIDIA H200 and a
+MacBook Pro (M4, 16 GB).
 ## Usage
+Install llama.cpp, which includes `llama-server`, following its
+[installation guide](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md) (Homebrew,
+winget, conda, Nix, and builds for NVIDIA GPUs).
+Then start a server; it downloads the model on first run and serves an OpenAI-compatible API and a chat
+page at `http://localhost:8080`:
 ```bash
+# Full thinking budget (needs about 15 GB of GPU memory)
 llama-server -hf FirstPrinciplesLabs/quark-GGUF:Q8_0 --jinja -ngl 99 -c 135168
+# 16 GB Macs: 32K context (longer answers are cut off)
 llama-server -hf FirstPrinciplesLabs/quark-GGUF:Q8_0 --jinja -ngl 99 -c 32768
 ```
+Use the sampling settings from the [Quark model card](https://huggingface.co/FirstPrinciplesLabs/quark#usage)
 (thinking mode: temperature 1.0, top_p 0.95, top_k 20, presence penalty 1.5).
 ## License
 Apache 2.0