Instructions to use Lucebox/gemma-4-31B-it-DFlash-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Lucebox/gemma-4-31B-it-DFlash-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Lucebox/gemma-4-31B-it-DFlash-GGUF",
	filename="gemma-4-31B-it-DFlash-q8_0.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Lucebox/gemma-4-31B-it-DFlash-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0
# Run inference directly in the terminal:
llama-cli -hf Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0
# Run inference directly in the terminal:
llama-cli -hf Lucebox/gemma-4-31B-it-DFlash-GGUF: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 Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf Lucebox/gemma-4-31B-it-DFlash-GGUF: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 Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0

Use Docker

docker model run hf.co/Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0

LM Studio
Jan
Ollama
How to use Lucebox/gemma-4-31B-it-DFlash-GGUF with Ollama:
```
ollama run hf.co/Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0
```

Unsloth Studio new

How to use Lucebox/gemma-4-31B-it-DFlash-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 Lucebox/gemma-4-31B-it-DFlash-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 Lucebox/gemma-4-31B-it-DFlash-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Lucebox/gemma-4-31B-it-DFlash-GGUF to start chatting

Docker Model Runner
How to use Lucebox/gemma-4-31B-it-DFlash-GGUF with Docker Model Runner:
```
docker model run hf.co/Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0
```

Lemonade

How to use Lucebox/gemma-4-31B-it-DFlash-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Lucebox/gemma-4-31B-it-DFlash-GGUF:Q8_0

Run and chat with the model

lemonade run user.gemma-4-31B-it-DFlash-GGUF-Q8_0

List all available models

lemonade list

davide221 commited on 1 day ago

Commit

9a280bb

verified ·

1 Parent(s): c4dff4c

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +59 -0

README.md ADDED Viewed

	@@ -0,0 +1,59 @@

+---
+license: apache-2.0
+library_name: gguf
+base_model: z-lab/gemma-4-31B-it-DFlash
+tags:
+  - gemma
+  - gemma-4
+  - dflash
+  - speculative-decoding
+  - block-diffusion
+  - draft-model
+  - gguf
+  - q8_0
+---
+# gemma-4-31B-it DFlash Draft — Q8_0 GGUF
+Q8_0 GGUF quantization of the [z-lab/gemma-4-31B-it-DFlash](https://huggingface.co/z-lab/gemma-4-31B-it-DFlash) draft model, produced for the [Lucebox dflash engine](https://github.com/Luce-Org/lucebox-hub) (speculative decoding for `google/gemma-4-31B-it`).
+- **Source**: [z-lab/gemma-4-31B-it-DFlash](https://huggingface.co/z-lab/gemma-4-31B-it-DFlash) (BF16 safetensors)
+- **Tool**: `quantize_gemma_dflash_q8.py` (parameterized variant of `dflash/scripts/quantize_draft_q8.py`)
+- **Size**: 1.63 GB (53% of BF16)
+- **Arch**: `gemma4-dflash-draft`
+- **Layers**: 5
+- **Hidden**: 5376
+- **n_head / n_head_kv**: 64 / 8
+- **head_dim**: 128
+- **intermediate_size**: 10752
+- **vocab_size**: 262144
+- **rope_theta**: 1e6, **rms_eps**: 1e-6
+- **sliding_window**: 2048, **final_logit_softcapping**: 30.0
+- **DFlash**: `n_target_layers=60`, `target_layer_ids=[1,12,23,35,46,57]`, `block_size=16`, `mask_token_id=4`
+- **Tensors**: projection weights → Q8_0, norm weights → F32 (precision-critical, tiny)
+## Pairs with
+- Target: [google/gemma-4-31B-it](https://huggingface.co/google/gemma-4-31B-it) (run as Q4_K_M GGUF via `unsloth/gemma-4-31B-it-GGUF` or `bartowski/google_gemma-4-31B-it-GGUF`)
+## Notes
+- The dflash-specific singletons `dflash.fc.weight` and `dflash.hidden_norm.weight` bridge target hidden states into the draft. Do **not** re-quantize with stock `llama-quantize` — it strips these tensors. Use the script above.
+- Loader support for `gemma4-dflash-draft` in `lucebox-hub` is the next step after PR #232 (gemma4 target adapter).
+## Usage
+```bash
+# With dflash_server (once gemma4-dflash-draft arch is wired in the loader)
+dflash_server gemma-4-31B-it-Q4_K_M.gguf --draft gemma-4-31B-it-DFlash-q8_0.gguf
+```
+## Conversion command
+```bash
+PYTHONPATH=lucebox-hub/dflash/deps/llama.cpp/gguf-py \
+python3 quantize_gemma_dflash_q8.py \
+  gemma-4-31B-it-DFlash/ \
+  gemma-4-31B-it-DFlash-q8_0.gguf \
+  --name gemma-4-31B-it-DFlash-Q8_0
+```