Instructions to use Johnblick187/gemma3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Johnblick187/gemma3 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Johnblick187/gemma3",
	filename="gemma-3-27b-it-abliterated.BF16-00002-of-00002.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Johnblick187/gemma3 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Johnblick187/gemma3:BF16
# Run inference directly in the terminal:
llama-cli -hf Johnblick187/gemma3:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Johnblick187/gemma3:BF16
# Run inference directly in the terminal:
llama-cli -hf Johnblick187/gemma3:BF16

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 Johnblick187/gemma3:BF16
# Run inference directly in the terminal:
./llama-cli -hf Johnblick187/gemma3:BF16

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 Johnblick187/gemma3:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Johnblick187/gemma3:BF16

Use Docker

docker model run hf.co/Johnblick187/gemma3:BF16

LM Studio
Jan
Ollama
How to use Johnblick187/gemma3 with Ollama:
```
ollama run hf.co/Johnblick187/gemma3:BF16
```

Unsloth Studio new

How to use Johnblick187/gemma3 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 Johnblick187/gemma3 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 Johnblick187/gemma3 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Johnblick187/gemma3 to start chatting

Docker Model Runner
How to use Johnblick187/gemma3 with Docker Model Runner:
```
docker model run hf.co/Johnblick187/gemma3:BF16
```

Lemonade

How to use Johnblick187/gemma3 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Johnblick187/gemma3:BF16

Run and chat with the model

lemonade run user.gemma3-BF16

List all available models

lemonade list

gemma3 / config.json

Johnblick187

Trained with Unsloth - config

94fe3ce verified about 2 months ago

raw

history blame contribute delete

3.95 kB

	{
	"architectures": [
	"Gemma3ForConditionalGeneration"
	],
	"boi_token_index": 255999,
	"bos_token_id": 2,
	"torch_dtype": "bfloat16",
	"eoi_token_index": 256000,
	"eos_token_id": 1,
	"image_token_index": 262144,
	"initializer_range": 0.02,
	"mm_tokens_per_image": 256,
	"model_name": "mlabonne/gemma-3-27b-it-abliterated",
	"model_type": "gemma3",
	"pad_token_id": 0,
	"text_config": {
	"_sliding_window_pattern": 6,
	"attention_bias": false,
	"attention_dropout": 0.0,
	"attn_logit_softcapping": null,
	"cache_implementation": "hybrid",
	"torch_dtype": "bfloat16",
	"final_logit_softcapping": null,
	"head_dim": 128,
	"hidden_activation": "gelu_pytorch_tanh",
	"hidden_size": 5376,
	"initializer_range": 0.02,
	"intermediate_size": 21504,
	"layer_types": [
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention"
	],
	"max_position_embeddings": 131072,
	"model_type": "gemma3_text",
	"num_attention_heads": 32,
	"num_hidden_layers": 62,
	"num_key_value_heads": 16,
	"query_pre_attn_scalar": 168,
	"rms_norm_eps": 1e-06,
	"rope_local_base_freq": 10000.0,
	"rope_scaling": {
	"factor": 8.0,
	"rope_type": "linear"
	},
	"rope_theta": 1000000.0,
	"sliding_window": 1024,
	"use_cache": true,
	"vocab_size": 262208
	},
	"unsloth_version": "2026.4.4",
	"vision_config": {
	"attention_dropout": 0.0,
	"torch_dtype": "bfloat16",
	"hidden_act": "gelu_pytorch_tanh",
	"hidden_size": 1152,
	"image_size": 896,
	"intermediate_size": 4304,
	"layer_norm_eps": 1e-06,
	"model_type": "siglip_vision_model",
	"num_attention_heads": 16,
	"num_channels": 3,
	"num_hidden_layers": 27,
	"patch_size": 14,
	"vision_use_head": false
	}
	}