Instructions to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF",
	filename="GLM-4.7-Flash-Uncen-Hrt-NEO-CODE-MAX-imat-D_AU-IQ2_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-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 DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-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 DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M

Use Docker

docker model run hf.co/DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-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": "DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M

Ollama
How to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF with Ollama:
```
ollama run hf.co/DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M
```

Unsloth Studio

How to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-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 DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-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 DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF to start chatting

How to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-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 DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF with Docker Model Runner:
```
docker model run hf.co/DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M
```

Lemonade

How to use DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF-Q4_K_M

List all available models

lemonade list

Hoping someone can finally help me with this

by ElvisM - opened Apr 19

Discussion

ElvisM

Apr 19

Anybody else here having the same problem where the model works well for the first message, and then it instantly starts to give gibberish for no reason in the second message? It looks like such a weird bug with llama.cpp. It happens with all GLM 4.7 models, which is sad because it's such a good LLM. It looks like reloading fixes it, but I'm surprised this model's implementation is still broken.

DavidAU

Owner Apr 20

Quant? Ai APP used? Settings ?

ElvisM

Apr 23

Quant? Ai APP used? Settings ?

Hi, I'm using Q4KM, Oobabooga's TextGen Web UI, and some GLM 4.7's settings that I found for SillyTavern somewhere on Google. It has repetition penalty disabled, as is recommended.

DavidAU

Owner Apr 24

Testing was done in LMSTudio ; if memory serves you can select the template to use in "TextGen".
This is CRITICAL.

RE: Settings; start with GLM 4.7 Flash settings/samplers.
Silly tavern sampler - there are tonnes of these - may be the root issue ; specially "Dry", and other advanced ones.

ElvisM

Apr 30

Testing was done in LMSTudio ; if memory serves you can select the template to use in "TextGen".
This is CRITICAL.

RE: Settings; start with GLM 4.7 Flash settings/samplers.
Silly tavern sampler - there are tonnes of these - may be the root issue ; specially "Dry", and other advanced ones.

Thanks. I've also tested the model only in Oobabooga (no SillyTavern) and the same problem still occurred. I've stopped using it for now because of this, so I don't know if anything has changed. Oobabooga just uses whatever template already comes bundled with the model. I just find it so weird that the first few messages are so good and then the model breaks, and reloading it seems to fix it.

DavidAU

Owner Apr 30

Based on your description; this could be a template glitch and/or a caching issue.
Older models are a lot more fault tolerant with Jinja template issues; where as newer ones can "drop the ball" if a few tokens are added at the wrong time.

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