Instructions to use alimpfard/diffusiongemma-ft-grammar with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use alimpfard/diffusiongemma-ft-grammar with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="alimpfard/diffusiongemma-ft-grammar",
	filename="diffusiongemma-grammar-Q8_0.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use alimpfard/diffusiongemma-ft-grammar 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 alimpfard/diffusiongemma-ft-grammar:Q8_0
# Run inference directly in the terminal:
llama cli -hf alimpfard/diffusiongemma-ft-grammar:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf alimpfard/diffusiongemma-ft-grammar:Q8_0
# Run inference directly in the terminal:
llama cli -hf alimpfard/diffusiongemma-ft-grammar:Q8_0

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 alimpfard/diffusiongemma-ft-grammar:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf alimpfard/diffusiongemma-ft-grammar:Q8_0

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 alimpfard/diffusiongemma-ft-grammar:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf alimpfard/diffusiongemma-ft-grammar:Q8_0

Use Docker

docker model run hf.co/alimpfard/diffusiongemma-ft-grammar:Q8_0

LM Studio
Jan
Ollama
How to use alimpfard/diffusiongemma-ft-grammar with Ollama:
```
ollama run hf.co/alimpfard/diffusiongemma-ft-grammar:Q8_0
```

Unsloth Studio

How to use alimpfard/diffusiongemma-ft-grammar 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 alimpfard/diffusiongemma-ft-grammar 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 alimpfard/diffusiongemma-ft-grammar to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for alimpfard/diffusiongemma-ft-grammar to start chatting

How to use alimpfard/diffusiongemma-ft-grammar with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf alimpfard/diffusiongemma-ft-grammar:Q8_0

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": "alimpfard/diffusiongemma-ft-grammar:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use alimpfard/diffusiongemma-ft-grammar with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf alimpfard/diffusiongemma-ft-grammar:Q8_0

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 alimpfard/diffusiongemma-ft-grammar:Q8_0

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use alimpfard/diffusiongemma-ft-grammar with Docker Model Runner:
```
docker model run hf.co/alimpfard/diffusiongemma-ft-grammar:Q8_0
```

Lemonade

How to use alimpfard/diffusiongemma-ft-grammar with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull alimpfard/diffusiongemma-ft-grammar:Q8_0

Run and chat with the model

lemonade run user.diffusiongemma-ft-grammar-Q8_0

List all available models

lemonade list

alimpfard commited on 9 days ago

Commit

dfc2869

verified ·

1 Parent(s): 7463f88

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -8

README.md CHANGED Viewed

@@ -7,12 +7,6 @@ tags:
 ---
 # diffusiongemma-ft-grammar
-Grammar / punctuation / markdown corrector finetuned from `unsloth/diffusiongemma-26B-A4B-it`
-(block-diffusion 26B-A4B MoE) via SFT on LanguageTool-corrected targets (with a proper-noun guard).
-Behavior: fixes grammar, punctuation, tense and broken markdown; preserves the author's wording and
-word count (~5%); returns only the corrected text; ignores prompt-injections embedded in the input.
-- **Full** bf16 merged weights — repo root safetensors.
-- **Q8_0 GGUF**: `diffusiongemma-grammar-Q8_0.gguf` — run with the diffusion-gemma llama.cpp build
-  (`llama-diffusion-cli` / `diffusion-gemma-eval`).

 ---
 # diffusiongemma-ft-grammar
+Grammar / punctuation / markdown corrector finetuned from `unsloth/diffusiongemma-26B-A4B-it`.
+Fixes grammar, punctuation, tense and broken markdown; preserves the author's wording and word count (~5%); returns only the corrected text; Prompt by `Correct grammar.\n<input text>`.