Instructions to use ertghiu256/gemma-4-e2b-deepglm-distill-chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ertghiu256/gemma-4-e2b-deepglm-distill-chat with Transformers:

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

pipe = pipeline("image-text-to-text", model="ertghiu256/gemma-4-e2b-deepglm-distill-chat")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("ertghiu256/gemma-4-e2b-deepglm-distill-chat")
model = AutoModelForMultimodalLM.from_pretrained("ertghiu256/gemma-4-e2b-deepglm-distill-chat", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ertghiu256/gemma-4-e2b-deepglm-distill-chat with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ertghiu256/gemma-4-e2b-deepglm-distill-chat"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ertghiu256/gemma-4-e2b-deepglm-distill-chat",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/ertghiu256/gemma-4-e2b-deepglm-distill-chat

SGLang

How to use ertghiu256/gemma-4-e2b-deepglm-distill-chat 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 "ertghiu256/gemma-4-e2b-deepglm-distill-chat" \
    --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": "ertghiu256/gemma-4-e2b-deepglm-distill-chat",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "ertghiu256/gemma-4-e2b-deepglm-distill-chat" \
        --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": "ertghiu256/gemma-4-e2b-deepglm-distill-chat",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use ertghiu256/gemma-4-e2b-deepglm-distill-chat with Docker Model Runner:
```
docker model run hf.co/ertghiu256/gemma-4-e2b-deepglm-distill-chat
```

Gemma-4-E2B-DeepGLM-Chat

Gemma-4-E2B-DeepGLM-Chat is a specialized, lightweight reasoning model fine-tuned on top of Google Gemma 4 E2B (unsloth/gemma-4-E2B-it). Designed to punch far above its parameter weight class (~2.3B active parameters), this model merges edge-optimized efficiency with advanced chain-of-thought (CoT) tracking, mathematical verification, and code execution capabilities.

🌟 Key Features

Efficient Architecture: Built on the Google Gemma 4 E2B edge architecture, optimized for low latency and minimal memory overhead.
Multimodal Edge Foundations: Inherits text, vision, and audio processing capabilities natively from the base Gemma 4 E2B lineup.
Deep CoT Reasoning: Enhanced logical tracking enabled via structured internal reasoning channels (<|channel>thought).
Targeted Fine-Tuning Datasets:
- isaiahbjork/cot-logic-reasoning (Formal logic, fallacies, multi-step deduction)
- Banaxi-Tech/Deepseek-V4-Reasoning-Code-2500 (Deterministic code execution & algorithm tracing)
- best-distill/glm-5.3-flash-distillation-chat (Mathematical distillation & structured formatting)
- armand0e/Fable-5-Chat (Conversational multi-turn engagement & technical instruction)

🚀 Quickstart

Using `transformers`

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "ertghiu256/gemma-4-e2b-deepglm-distill-chat"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)

prompt = "Why is the sky blue?"

inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512, temperature=0.2)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))

💡 Output Format

To get the absolute best reasoning quality from this model, allow the model to utilize its built-in reasoning tags:

<|channel>thought
[Internal reasoning trace / scratchpad]
<channel|>
[Final response]

Tip: Running it with a moderate temperature (0.5 - 0.9).

📄 Citation & Acknowledgments

Base Model: Google Gemma 4 E2B (google/gemma-4-e2b)
Fine-Tuning Datasets:
Banaxi-Tech/Deepseek-V4-Reasoning-Code-2500
best-distill/glm-5.3-flash-distillation-chat
armand0e/Fable-5-Chat
isaiahbjork/cot-logic-reasoning

Citation

@misc{gemmateam2026gemma4,
      title={Gemma 4 Technical Report}, 
      author={Gemma Team},
      year={2026},
      eprint={2607.02770},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2607.02770}, 
}

Uploaded finetuned

Developed by: ertghiu256
License: apache-2.0
Finetuned from model : unsloth/gemma-4-E2B-it

This gemma4 model was trained 2x faster with Unsloth and Huggingface's TRL library.

Downloads last month: 40

Safetensors

Model size

5B params

Tensor type

BF16

Model tree for ertghiu256/gemma-4-e2b-deepglm-distill-chat

Base model

google/gemma-4-E2B

Finetuned

google/gemma-4-E2B-it

Finetuned

(424)

this model

Datasets used to train ertghiu256/gemma-4-e2b-deepglm-distill-chat

Paper for ertghiu256/gemma-4-e2b-deepglm-distill-chat

Gemma 4 Technical Report

Paper • 2607.02770 • Published Jul 2 • 86