Instructions to use Abiray/Hemmingway-1-heretic-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 Abiray/Hemmingway-1-heretic-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 Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Abiray/Hemmingway-1-heretic-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 Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Abiray/Hemmingway-1-heretic-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 Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Abiray/Hemmingway-1-heretic-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 Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use Abiray/Hemmingway-1-heretic-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Abiray/Hemmingway-1-heretic-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Abiray/Hemmingway-1-heretic-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M

Ollama
How to use Abiray/Hemmingway-1-heretic-GGUF with Ollama:
```
ollama run hf.co/Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M
```
Unsloth Desktop

How to use Abiray/Hemmingway-1-heretic-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M

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": "Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Abiray/Hemmingway-1-heretic-GGUF with Docker Model Runner:
```
docker model run hf.co/Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M
```

Lemonade

How to use Abiray/Hemmingway-1-heretic-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Hemmingway-1-heretic-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use Abiray/Hemmingway-1-heretic-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 Abiray/Hemmingway-1-heretic-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 Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Abiray/Hemmingway-1-heretic-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Abiray/Hemmingway-1-heretic-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 "Abiray/Hemmingway-1-heretic-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"

Abiray commited on 8 days ago

Commit

32203c0

verified ·

1 Parent(s): 4119fa5

Update README.md

Browse files

Files changed (1) hide show

README.md +49 -34

README.md CHANGED Viewed

@@ -1,60 +1,75 @@
 ---
 base_model: Altworld/Hemmingway-1
 tags:
 - gguf
 - llama.cpp
-- qwen35
-- conversational
-- text-generation
-- mtp
 pipeline_tag: text-generation
-license: apache-2.0
 language:
 - en
 ---
 # Hemmingway-1-heretic-GGUF
-High-fidelity **GGUF** quantizations of `arelath/Hemmingway-1-heretic-adapter` merged into `Altworld/Hemmingway-1` in `bfloat16`.
-**These builds preserve the full 65-layer architecture, including the 65th Multi-Token Prediction (MTP) speculative drafting head.**
-## Models Available
-| Quantization | File | Size | Notes |
-| :--- | :--- | :--- | :--- |
-| **Q3_K_M** | `Hemmingway-1-heretic-Q3_K_M.gguf` | 12.57 GB | Full 65-layer backbone + MTP head |
-| **Q4_K_M** | `Hemmingway-1-heretic-Q4_K_M.gguf` | 15.66 GB | Full 65-layer backbone + MTP head |
-| **Q4_K_S** | `Hemmingway-1-heretic-Q4_K_S.gguf` | 14.74 GB | Full 65-layer backbone + MTP head |
-| **Q5_K_M** | `Hemmingway-1-heretic-Q5_K_M.gguf` | 18.19 GB | Full 65-layer backbone + MTP head |
-| **Q6_K** | `Hemmingway-1-heretic-Q6_K.gguf` | 20.89 GB | Full 65-layer backbone + MTP head |
-| **Q8_0** | `Hemmingway-1-heretic-Q8_0.gguf` | 27.05 GB | Full 65-layer backbone + MTP head |
 ---
-## Speculative Decoding (MTP Acceleration)
-Enable native speculative lookahead decoding in `llama.cpp` using the preserved draft head:
-```bash
-llama-cli -m Hemmingway-1-heretic-Q4_K_M.gguf --spec-type draft-mtp -p "Hello!"
-```
-## Quick Start with llama.cpp
-### Local Server:
-```bash
-llama-server -hf Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M --spec-type draft-mtp --port 8080 -t 16
-```
-### Direct CLI Chat:
 ```bash
-llama-cli -hf Abiray/Hemmingway-1-heretic-GGUF:Q4_K_M --spec-type draft-mtp -cnv -t 16
 ```
----
-## Lineage & Provenance
-* **Base Architecture:** [Altworld/Hemmingway-1](https://huggingface.co/Altworld/Hemmingway-1)
-* **Adapter:** [arelath/Hemmingway-1-heretic-adapter](https://huggingface.co/arelath/Hemmingway-1-heretic-adapter)
-* **Quantization Engine:** `llama.cpp` (16-thread native build)

 ---
+license: cc-by-nc-4.0
 base_model: Altworld/Hemmingway-1
 tags:
+- text-generation
 - gguf
 - llama.cpp
+- qwen3.8
+- chat
+- creative-writing
+- uncensored
+- heretic
+- abliterated
+- decensored
 pipeline_tag: text-generation
 language:
 - en
 ---
 # Hemmingway-1-heretic-GGUF
+This repository contains static GGUF quantizations of **Hemmingway-1-heretic**, created by applying the [arelath/Hemmingway-1-heretic-adapter](https://huggingface.co/arelath/Hemmingway-1-heretic-adapter) to [Altworld/Hemmingway-1](https://huggingface.co/Altworld/Hemmingway-1) (a 27B model built on Qwen3.8-27B) and quantizing for local inference with `llama.cpp`.
+Hemmingway-1 is tuned specifically to write like a real person—skipping preambles, lecture notes, and corporate filler. This **heretic** version decensors the weights via targeted representation abliteration using Heretic v2.0.0, reducing refusals to near-zero while preserving high-EQ conversational voice.
+---
+## Quantization Variants
+| Filename | Size | Recommended RAM/VRAM | Description |
+| :--- | :---: | :---: | :--- |
+| `Hemmingway-1-heretic-Q3_K_M.gguf` | 13.5 GB | ~16 GB | Maximum compression; fits on mid-range hardware with slight perplexity trade-off. |
+| `Hemmingway-1-heretic-Q4_K_S.gguf` | 15.8 GB | ~19 GB | Lightweight 4-bit; fast tokens/sec for tighter memory envelopes. |
+| `Hemmingway-1-heretic-Q4_K_M.gguf` | 16.8 GB | ~20 GB | **Recommended balance.** Retains near-full baseline accuracy with solid speed. |
+| `Hemmingway-1-heretic-Q5_K_M.gguf` | 19.5 GB | ~23 GB | High fidelity; preserves subtle nuanced phrasing and tone. |
+| `Hemmingway-1-heretic-Q6_K.gguf`   | 22.4 GB | ~26 GB | Near loss-less reproduction of the 16-bit unquantized model. |
+| `Hemmingway-1-heretic-Q8_0.gguf`   | 29.0 GB | ~33 GB | Maximum quality 8-bit quantization for high-VRAM setups. |
 ---
+## Decensoring & Abliteration Details
+Created using [Heretic](https://heretic-project.org) v2.0.0.dev0 against `Altworld/Hemmingway-1`.
+### Abliteration Parameters
+- **direction_index:** 38.05
+- **attn.o_proj.max_weight:** 1.03 (position: 41.01, min_weight: 1.02, min_distance: 28.56)
+- **mlp.down_proj.max_weight:** 1.41 (position: 51.07, min_weight: 1.39, min_distance: 35.29)
+### Refusal Comparison
+| Metric | This Model (Heretic) | Original Hemmingway-1 |
+| :--- | :---: | :---: |
+| **Refusals** | **2 / 100** | 95 / 100 |
+| **KL Divergence** | **0.1662** | 0 *(baseline)* |
+---
+## Quickstart & Usage
+### 1. llama.cpp CLI
 ```bash
+# Clone and build llama.cpp, then run:
+./llama-cli \
+  -m Hemmingway-1-heretic-Q4_K_M.gguf \
+  -p "<|im_start|>system\nYou are a direct, empathetic writing assistant.<|im_end|>\n<|im_start|>user\nWrite the text I send my landlord about the broken boiler.<|im_end|>\n<|im_start|>assistant\n" \
+  -n 512 \
+  -ngl 99 \
+  -c 8192
 ```
+## Credits & Licensing
+**Base Weights:** Altworld/Hemmingway-1 by Altworld, built on Qwen/Qwen3.8-27B.
+**Abliteration Adapter:** arelath/Hemmingway-1-heretic-adapter by arelath.
+**License:** Governed by the original model's license terms: CC BY-NC 4.0 (Free for personal and non-commercial research use; commercial usage requires agreement with Altworld).