Instructions to use yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF",
	filename="MTP/gemma-4-12B-it-MTP-BF16.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 yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-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 yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-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 yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-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 yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-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 yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M

Use Docker

docker model run hf.co/yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-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": "yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M

Ollama
How to use yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF with Ollama:
```
ollama run hf.co/yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M
```

Unsloth Studio

How to use yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-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 yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-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 yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF to start chatting

How to use yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-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": "yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-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 yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF with Docker Model Runner:
```
docker model run hf.co/yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M
```

Lemonade

How to use yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF-Q4_K_M

List all available models

lemonade list

yuxinlu1 commited on 12 days ago

Commit

190a313

verified ·

1 Parent(s): e8a1d38

Add Speculative decoding (MTP draft) section: verified build llama.cpp b9553 (9e3b928fd), regression note for b9702/b9717

Browse files

Files changed (1) hide show

README.md +29 -0

README.md CHANGED Viewed

@@ -207,3 +207,32 @@ handles it). Recommended sampling: `temp 1.0, top_p 0.95, top_k 64`; for coding
   **Apache 2.0** too — free to use, modify, and redistribute. 🎉
 - **Base model:** [`google/gemma-4-12B-it`](https://huggingface.co/google/gemma-4-12B-it).
 - Personal/hobby project — shared as-is, no warranty. Built with time, care, and a lot of coffee. Have fun! 🐾✨

   **Apache 2.0** too — free to use, modify, and redistribute. 🎉
 - **Base model:** [`google/gemma-4-12B-it`](https://huggingface.co/google/gemma-4-12B-it).
 - Personal/hobby project — shared as-is, no warranty. Built with time, care, and a lot of coffee. Have fun! 🐾✨
+---
+## ⚡ Speculative decoding (MTP draft) — verified build
+The `MTP/` folder ships the Gemma 4 multi-token-prediction draft (unsloth's GGUF conversion of Google's official
+`gemma-4-12B-it-assistant`) for speculative decoding. Gemma 4 MTP is in **llama.cpp mainline** (PR #23398) — no fork
+needed — but the `gemma4-assistant` loader is **build-sensitive right now**, so please use the exact build below:
+- ✅ **Verified working: llama.cpp `b9553` (commit `9e3b928fd`).** I reproduced it with `gemma4-v2-Q8_0` + the
+  `MTP-Q8_0` draft: loads cleanly and accelerates generation (~88 → ~180 tok/s on a simple deterministic prompt;
+  expect ~1.2–1.3× on real coding/thinking). **Lossless** either way.
+- ⚠️ **Newer builds (e.g. b9702 / b9717) currently crash** while loading the draft with `invalid vector subscript`.
+  This is an **upstream regression** in the `gemma4-assistant` loader path, *not* a problem with these GGUFs — the same
+  files load fine on b9553. Stick with **b9553** until it's fixed upstream.
+Working command on b9553 (note the older flag names — `--model-draft`, **not** `--spec-draft-model`):
+```bat
+llama-server -m gemma4-v2-Q8_0.gguf ^
+  --model-draft MTP\gemma-4-12B-it-MTP-Q8_0.gguf ^
+  --spec-type draft-mtp --spec-draft-n-max 4 ^
+  -ngl 99 -ngld 99 -fa on --jinja
+```
+> ℹ️ The `Gemma4Assistant requires ctx_other to be set (this is normal during memory fitting)` line is harmless. The
+> draft is the generic Gemma 4 assistant (not retrained for v2), so acceptance is a touch lower than a model-specific
+> draft would give — still 100% lossless. On small-VRAM cards, Q8 main + long context + the draft can be tight; drop to
+> Q6_K/Q4_K_M or a smaller `--ctx-size` if you hit OOM.