Instructions to use LLMWildling/gemma4-39b-a6b-preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LLMWildling/gemma4-39b-a6b-preview with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="LLMWildling/gemma4-39b-a6b-preview")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForCausalLM

processor = AutoProcessor.from_pretrained("LLMWildling/gemma4-39b-a6b-preview")
model = AutoModelForCausalLM.from_pretrained("LLMWildling/gemma4-39b-a6b-preview")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use LLMWildling/gemma4-39b-a6b-preview with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LLMWildling/gemma4-39b-a6b-preview"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLMWildling/gemma4-39b-a6b-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LLMWildling/gemma4-39b-a6b-preview

SGLang

How to use LLMWildling/gemma4-39b-a6b-preview with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "LLMWildling/gemma4-39b-a6b-preview" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLMWildling/gemma4-39b-a6b-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "LLMWildling/gemma4-39b-a6b-preview" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLMWildling/gemma4-39b-a6b-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LLMWildling/gemma4-39b-a6b-preview with Docker Model Runner:
```
docker model run hf.co/LLMWildling/gemma4-39b-a6b-preview
```

Gemma4-39B-A6B Preview

gemma4-39b-a6b-preview is a Gemma4 MoE expansion of unsloth/Gemma-4-26B-A4B-it, built with an internal pre-training/post-training pipeline targeting software engineering, repository reasoning, agentic workflows, and general instruction following.

This preview build was produced on a 2-GPU local training setup. A larger 52B build is planned to follow soon as the more stable release line.

Model Details

Base model: unsloth/Gemma-4-26B-A4B-it
Architecture: Gemma4 MoE
Total logical parameters: approximately 38.7B
Active parameters: approximately 5.9B at the default active expert budget
Expert layout: 128 base experts + 64 selected specialist experts
Context target: up to 131k tokens in vLLM serving
Primary focus: SWE, code/repository analysis, agentic traces, and reasoning
Recommended temperature: 0.0 to 0.7 for agentic/tool use

Serving

Use the included Gemma4 chat template. Thinking should be enabled for best agentic behavior.

CUDA_VISIBLE_DEVICES=0 \
VLLM_ALLOW_INSECURE_SERIALIZATION=1 \
vllm serve . \
  --served-model-name gemma4-39b-a6b-preview \
  --host 0.0.0.0 \
  --port 23333 \
  --max-model-len 131072 \
  --tensor-parallel-size 1 \
  --gpu-memory-utilization 0.90 \
  --trust-remote-code \
  --chat-template ./chat_template.jinja \
  --default-chat-template-kwargs '{"enable_thinking": true}' \
  --enable-auto-tool-choice \
  --tool-call-parser gemma4 \
  --reasoning-parser gemma4

OpenAI-compatible endpoint:

http://localhost:23333/v1

Validation

Validated through Gemma4-native vLLM serving with thinking enabled across chat, SWE-style prompts, reasoning, parsed tool calls, and post-tool final answers.

Notes

This is a full model checkpoint, not a LoRA adapter.

Downloads last month: 17

Safetensors

Model size

22B params

Tensor type

F32

BF16

F8_E4M3