Instructions to use jialinyyzz/humanizer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jialinyyzz/humanizer with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="jialinyyzz/humanizer")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("jialinyyzz/humanizer")
model = AutoModelForMultimodalLM.from_pretrained("jialinyyzz/humanizer", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use jialinyyzz/humanizer 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 jialinyyzz/humanizer:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf jialinyyzz/humanizer:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf jialinyyzz/humanizer:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf jialinyyzz/humanizer: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 jialinyyzz/humanizer:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf jialinyyzz/humanizer: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 jialinyyzz/humanizer:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf jialinyyzz/humanizer:Q4_K_M

Use Docker

docker model run hf.co/jialinyyzz/humanizer:Q4_K_M

LM Studio
Jan

vLLM

How to use jialinyyzz/humanizer with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/jialinyyzz/humanizer:Q4_K_M

SGLang

How to use jialinyyzz/humanizer with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "jialinyyzz/humanizer" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jialinyyzz/humanizer",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "jialinyyzz/humanizer" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jialinyyzz/humanizer",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use jialinyyzz/humanizer with Ollama:
```
ollama run hf.co/jialinyyzz/humanizer:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use jialinyyzz/humanizer with Docker Model Runner:
```
docker model run hf.co/jialinyyzz/humanizer:Q4_K_M
```

Lemonade

How to use jialinyyzz/humanizer with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull jialinyyzz/humanizer:Q4_K_M

Run and chat with the model

lemonade run user.humanizer-Q4_K_M

List all available models

lemonade list

Atomic Chat

Start here (v2, updated Oct 9): app, file sizes, prompt format, what feedback helps

by jialinyyzz - opened 25 days ago

Discussion

jialinyyzz

Owner 25 days ago

•

edited 2 days ago

Updated 2026-10-09 for v2 (12B). The September version of this post described an older model; its quantization advice no longer applies.

Four things that will save you time, then one ask.

1. Easiest: the app. Mac (Apple silicon) · Windows. It picks a model size for your memory, downloads it once and then works offline. In 0.3.2 you can import .docx and PDF, select any words in the draft and click Create fact so they come back exactly as written, and check for updates from the … menu (0.3.1 and earlier need one manual update). No install at all: the Space has the same Create fact and import.

2. Any file size works now, including 4-bit and 2-bit.

File	Size	Memory
`Q8_0` (recommended)	12.7 GB	32 GB+
`Q6_K`	10.0 GB	16 GB, no measurable loss
`Q4_K_M`	7.6 GB	~14 GB, slight loss
`Q3-QAT`	5.6 GB	12 GB
`IQ2_XS-QAT`	3.9 GB	8 GB

The Q4_K_M, Q3 and 2-bit files are trained to stay close to the full model, so they are not the gibberish you may remember from the old Q4. Q3 and 2-bit make a few more fact slips; check numbers and names. Full table: model card.

3. Prompt format. It is still a completion model trained on one exact wrapper (prompt_format.json, byte for byte). Since 2026-10-04 the GGUF files also carry a chat template that builds that wrapper from your last message, so llama-server's chat endpoint works and chat apps that use the file's template should too (one draft per message; system prompts are ignored). For scripts, agents and long documents use hz (USAGE.md).

4. Read the result before you send it. On 420 English rewrites a strict fact checker flagged a problem in 44, and about 9 in 10 of those were a single word, number or phrase, e.g. a dropped "about" or "estimated". The app highlights any number from the draft that is missing in the rewrite. Very long drafts (over about 1,000 tokens in one go) drift more; hz splits long documents into pieces for you.

What we would like back. Where does it break on your text? If it drops or changes a fact, keeps the AI tone, or the app misbehaves, post the draft and the output here (a single failing pair helps more than a general impression) and say which file or app version you used. Genres we cover thinly are especially useful: legal, medical, technical, non-English.

Code, app, evaluation set: https://github.com/sgaofen/humanizer-local-model

jialinyyzz changed discussion title from Start here: prompt format, what it gets wrong, and what feedback we need to Start here (v2, updated Oct 9): app, file sizes, prompt format, what feedback helps 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