Instructions to use heterodoxin/vibethinker-3b-apostate-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use heterodoxin/vibethinker-3b-apostate-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="heterodoxin/vibethinker-3b-apostate-gguf",
	filename="vibethinker-3b-apostate-Q2_K.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use heterodoxin/vibethinker-3b-apostate-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 heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf heterodoxin/vibethinker-3b-apostate-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 heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf heterodoxin/vibethinker-3b-apostate-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 heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf heterodoxin/vibethinker-3b-apostate-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 heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M

Use Docker

docker model run hf.co/heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M

LM Studio
Jan
Ollama
How to use heterodoxin/vibethinker-3b-apostate-gguf with Ollama:
```
ollama run hf.co/heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M
```

Unsloth Studio

How to use heterodoxin/vibethinker-3b-apostate-gguf 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 heterodoxin/vibethinker-3b-apostate-gguf 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 heterodoxin/vibethinker-3b-apostate-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for heterodoxin/vibethinker-3b-apostate-gguf to start chatting

Atomic Chat new
Docker Model Runner
How to use heterodoxin/vibethinker-3b-apostate-gguf with Docker Model Runner:
```
docker model run hf.co/heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M
```

Lemonade

How to use heterodoxin/vibethinker-3b-apostate-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M

Run and chat with the model

lemonade run user.vibethinker-3b-apostate-gguf-Q4_K_M

List all available models

lemonade list

vibethinker-3b-apostate — GGUF

Join the community: Discord

GGUF quants of heterodoxin/vibethinker-3b-apostate, an Apostate weight-edit uncensored model. Pick a single quant or the full-quality bf16.

Quants

file	quant	notes
`vibethinker-3b-apostate-bf16.gguf`	BF16	full quality (default)
`vibethinker-3b-apostate-Q2_K.gguf`	Q2_K	smallest, lowest quality
`vibethinker-3b-apostate-Q3_K_M.gguf`	Q3_K_M	small
`vibethinker-3b-apostate-Q4_K_M.gguf`	Q4_K_M	recommended balance
`vibethinker-3b-apostate-Q5_K_M.gguf`	Q5_K_M	higher quality
`vibethinker-3b-apostate-Q6_K.gguf`	Q6_K	near-lossless
`vibethinker-3b-apostate-Q8_0.gguf`	Q8_0	highest quality quant

Install one quant (Ollama, easiest)

ollama run hf.co/heterodoxin/vibethinker-3b-apostate-gguf:Q4_K_M      # any quant tag above, or BF16

Install one quant (llama.cpp / manual)

huggingface-cli download heterodoxin/vibethinker-3b-apostate-gguf vibethinker-3b-apostate-Q4_K_M.gguf --local-dir .
./llama-cli -m vibethinker-3b-apostate-Q4_K_M.gguf -p "your prompt"

Swap Q4_K_M for any quant in the table, or vibethinker-3b-apostate-bf16.gguf for full quality.

Downloads last month: 712

GGUF

Model size

3B params

Architecture

qwen2

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for heterodoxin/vibethinker-3b-apostate-gguf

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-Coder-3B

Finetuned

WeiboAI/VibeThinker-3B

Finetuned

heterodoxin/vibethinker-3b-apostate

Quantized

(3)

this model