Instructions to use thanglq150188/gemma-4-E2B-mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use thanglq150188/gemma-4-E2B-mini with Transformers:

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

pipe = pipeline("text-generation", model="thanglq150188/gemma-4-E2B-mini")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("thanglq150188/gemma-4-E2B-mini")
model = AutoModelForCausalLM.from_pretrained("thanglq150188/gemma-4-E2B-mini", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use thanglq150188/gemma-4-E2B-mini 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 thanglq150188/gemma-4-E2B-mini:Q8_0
# Run inference directly in the terminal:
llama cli -hf thanglq150188/gemma-4-E2B-mini:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf thanglq150188/gemma-4-E2B-mini:Q8_0
# Run inference directly in the terminal:
llama cli -hf thanglq150188/gemma-4-E2B-mini: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 thanglq150188/gemma-4-E2B-mini:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf thanglq150188/gemma-4-E2B-mini: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 thanglq150188/gemma-4-E2B-mini:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf thanglq150188/gemma-4-E2B-mini:Q8_0

Use Docker

docker model run hf.co/thanglq150188/gemma-4-E2B-mini:Q8_0

LM Studio
Jan

vLLM

How to use thanglq150188/gemma-4-E2B-mini with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "thanglq150188/gemma-4-E2B-mini"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "thanglq150188/gemma-4-E2B-mini",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/thanglq150188/gemma-4-E2B-mini:Q8_0

SGLang

How to use thanglq150188/gemma-4-E2B-mini 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 "thanglq150188/gemma-4-E2B-mini" \
    --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": "thanglq150188/gemma-4-E2B-mini",
		"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 "thanglq150188/gemma-4-E2B-mini" \
        --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": "thanglq150188/gemma-4-E2B-mini",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use thanglq150188/gemma-4-E2B-mini with Ollama:
```
ollama run hf.co/thanglq150188/gemma-4-E2B-mini:Q8_0
```

Unsloth Studio

How to use thanglq150188/gemma-4-E2B-mini 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 thanglq150188/gemma-4-E2B-mini 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 thanglq150188/gemma-4-E2B-mini to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for thanglq150188/gemma-4-E2B-mini to start chatting

How to use thanglq150188/gemma-4-E2B-mini with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf thanglq150188/gemma-4-E2B-mini: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": "thanglq150188/gemma-4-E2B-mini:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use thanglq150188/gemma-4-E2B-mini with Docker Model Runner:
```
docker model run hf.co/thanglq150188/gemma-4-E2B-mini:Q8_0
```

Lemonade

How to use thanglq150188/gemma-4-E2B-mini with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull thanglq150188/gemma-4-E2B-mini:Q8_0

Run and chat with the model

lemonade run user.gemma-4-E2B-mini-Q8_0

List all available models

lemonade list

Hermes Agent

How to use thanglq150188/gemma-4-E2B-mini with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf thanglq150188/gemma-4-E2B-mini: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 thanglq150188/gemma-4-E2B-mini:Q8_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use thanglq150188/gemma-4-E2B-mini with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf thanglq150188/gemma-4-E2B-mini:Q8_0

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "thanglq150188/gemma-4-E2B-mini:Q8_0" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Configuration Parsing Warning:In config.json: "num_experts" must be a number

Gemma 4 E2B Mini — text-only, vocabulary pruned to English + Vietnamese

A compact, vocabulary-pruned version of Gemma 4 E2B (text-only). The 262,144-token multilingual vocabulary is reduced to the 69,246 tokens used by English + Vietnamese, removing ~44% of the parameters with no measured quality loss on EN/VI. Built for efficient on-device finetuning and serving on a 12 GB GPU.

Built on Google's Gemma 4 (via principled-intelligence/gemma-4-E2B-it-text-only, the text-only checkpoint). Use is governed by the Gemma Terms of Use and the Gemma Prohibited Use Policy — https://ai.google.dev/gemma/terms . "Gemma is provided under and subject to the Gemma Terms of Use." This derivative keeps the Gemma name and license as required.

What changed vs. the base

	Base (text-only)	This model (Mini)
Vocab	262,144	69,246 (EN+VI)
Params	4.63 B	2.60 B
bf16 weights	9.26 GB	5.21 GB
q8_0 GGUF	4.6 GB	2.6 GB
EN/VI accuracy (UIT-VSFC, 300)	0.5567	0.5567 (Δ 0, 99.7% agreement)

59% of the base is vocab-dependent (token embeddings + Per-Layer Embeddings); these are not quantized during QLoRA, so pruning them is the highest-leverage size/VRAM win on small GPUs. It also raises the max QLoRA training context on a 12 GB card from ~1024 to ~7168 tokens.

How it was made

Keep-set = tokens used by an EN+VI corpus (VI+EN Wikipedia) + all 256 byte-fallback tokens + special tokens + every single EN/VI character token (69,246 total).
The BPE tokenizer is pruned (merges filtered so every kept merge stays applicable; ids remapped contiguously); byte-fallback remains, so any character is still encodable.
embed_tokens, embed_tokens_per_layer (PLE), and the tied lm_head are row-sliced to the keep-set. The original rows are kept verbatim, so EN/VI behaviour is unchanged — only the ability to emit other-language tokens is removed.

Usage (transformers)

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
ID = "thanglq150188/gemma-4-E2B-mini"
tok = AutoTokenizer.from_pretrained(ID)
m = AutoModelForCausalLM.from_pretrained(ID, dtype=torch.bfloat16, device_map="cuda")
msgs = [{"role": "user", "content": "Thủ đô của Việt Nam là thành phố nào?"}]
ids = tok.apply_chat_template(msgs, add_generation_prompt=True,
        return_tensors="pt", return_dict=True).to("cuda")
print(tok.decode(m.generate(**ids, max_new_tokens=64)[0, ids["input_ids"].shape[1]:],
                 skip_special_tokens=True))

Chat format is Gemma 4's <|turn> / <turn|> (not Gemma 2/3's <start_of_turn>).

Usage (llama.cpp / llama-server)

# gemma4-envi-pruned-q8_0.gguf and gemma4_chat.jinja are included in this repo
llama-server -m gemma4-envi-pruned-q8_0.gguf -ngl 99 -fa on \
  --parallel 8 --ctx-size 65536 --cont-batching \
  --chat-template-file gemma4_chat.jinja

~260 tok/s aggregate at 10 concurrent requests, ~3.0 GB VRAM (8×8192 ctx) on an RTX 3060. The model stays resident (no idle unload).

`vsfc-sentiment-lora/` — demo finetune (optional)

QLoRA adapters from finetuning this model on UIT-VSFC Vietnamese sentiment (4k examples). Lifts accuracy 0.68 → 0.95 on the 300-example test set. Load with peft.PeftModel.from_pretrained(base, "thanglq150188/gemma-4-E2B-mini", subfolder="vsfc-sentiment-lora"). Error analysis: 0 of 16 mistakes involved a pruned-away token — pruning is not the cause.

Limitations

English + Vietnamese only by design. Other languages/scripts now fragment into byte-fallback tokens and are not supported (that is the point — the vocab was trimmed).
Inherits the capabilities and limitations of the base Gemma 4 E2B model.

Downloads last month: 80

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for thanglq150188/gemma-4-E2B-mini

Base model

principled-intelligence/gemma-4-E2B-it-text-only

Quantized

(1)

this model

Gemma 4 E2B Mini — text-only, vocabulary pruned to English + Vietnamese

What changed vs. the base

How it was made

Usage (transformers)

Usage (llama.cpp / llama-server)

vsfc-sentiment-lora/ — demo finetune (optional)

Limitations

Model tree for thanglq150188/gemma-4-E2B-mini

`vsfc-sentiment-lora/` — demo finetune (optional)