Instructions to use mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF", dtype="auto")

llama-cpp-python

How to use mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF",
	filename="Huihui-GLM-4.7-Flash-abliterated-57B.i1-IQ1_M.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 mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-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 mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-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 mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-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 mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-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 mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF:Q4_K_M

Use Docker

docker model run hf.co/mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF with Ollama:
```
ollama run hf.co/mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF:Q4_K_M
```

Unsloth Studio

How to use mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-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 mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-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 mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF with Docker Model Runner:
```
docker model run hf.co/mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF:Q4_K_M
```

Lemonade

How to use mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull mradermacher/Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Huihui-GLM-4.7-Flash-abliterated-57B-i1-GGUF-Q4_K_M

List all available models

lemonade list

What is the purpose of this new 57B variant?

by sebastienbo - opened Jan 31

Discussion

sebastienbo

Jan 31

What is the purpose of this new 57B variant?

It's kind of weird because the official 4.7 flash model was 30B parameters, so where are the other parameters comming from?
And what kind Of differenc edoes that make in the active paramters? is it still 3B ?

nicoboss

Feb 2

Such models are usually made by self-merging the smaller model into a larger one. Doing so makes the model more intelligent but also more expensive to run but if someone really likes a certain model and has the resources to run a larger one then doing so is worth it. Here a popular example of a self-merge chain which resulted in FATLLAMA-1.7T-Instruct which to this day is the largest publicly available model on HuggingFace:

https://huggingface.co/RichardErkhov/FATLLAMA-1.7T-Instruct (self-merge of BigLlama-3.1-1T-Instruct)
https://huggingface.co/mlabonne/BigLlama-3.1-1T-Instruct (self-merge of BigLlama-3.1-681B-Instruct)
https://huggingface.co/mlabonne/BigLlama-3.1-681B-Instruct (self-merge of Llama-3.1-405B-Instruct)
https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct (base model)

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