Instructions to use mlx-community/gemma-3-27b-pt-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlx-community/gemma-3-27b-pt-8bit with Transformers:

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

pipe = pipeline("image-text-to-text", model="mlx-community/gemma-3-27b-pt-8bit")

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("mlx-community/gemma-3-27b-pt-8bit")
model = AutoModelForMultimodalLM.from_pretrained("mlx-community/gemma-3-27b-pt-8bit", device_map="auto")

MLX

How to use mlx-community/gemma-3-27b-pt-8bit with MLX:

# Make sure mlx-vlm is installed
# pip install --upgrade mlx-vlm

from mlx_vlm import load, generate
from mlx_vlm.prompt_utils import apply_chat_template
from mlx_vlm.utils import load_config

# Load the model
model, processor = load("mlx-community/gemma-3-27b-pt-8bit")
config = load_config("mlx-community/gemma-3-27b-pt-8bit")

# Prepare input
image = ["http://images.cocodataset.org/val2017/000000039769.jpg"]
prompt = "Describe this image."

# Apply chat template
formatted_prompt = apply_chat_template(
    processor, config, prompt, num_images=1
)

# Generate output
output = generate(model, processor, formatted_prompt, image)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

vLLM

How to use mlx-community/gemma-3-27b-pt-8bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlx-community/gemma-3-27b-pt-8bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlx-community/gemma-3-27b-pt-8bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mlx-community/gemma-3-27b-pt-8bit

SGLang

How to use mlx-community/gemma-3-27b-pt-8bit 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 "mlx-community/gemma-3-27b-pt-8bit" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlx-community/gemma-3-27b-pt-8bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "mlx-community/gemma-3-27b-pt-8bit" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlx-community/gemma-3-27b-pt-8bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mlx-community/gemma-3-27b-pt-8bit with Docker Model Runner:
```
docker model run hf.co/mlx-community/gemma-3-27b-pt-8bit
```
Atomic Chat

prince-canuma commited on Mar 18, 2025

Commit

82d19ea

verified ·

1 Parent(s): 8c4dead

Upload folder using huggingface_hub

Browse files

Files changed (10) hide show

README.md +1 -1
generation_config.json +8 -0
model-00001-of-00006.safetensors +2 -2
model-00002-of-00006.safetensors +2 -2
model-00003-of-00006.safetensors +2 -2
model-00004-of-00006.safetensors +2 -2
model-00005-of-00006.safetensors +2 -2
model-00006-of-00006.safetensors +2 -2
model.safetensors.index.json +0 -0
processor_config.json +4 -0

@@ -12,7 +12,7 @@ tags:
 ---
 # mlx-community/gemma-3-27b-pt-8bit
-This model was converted to MLX format from [`google/gemma-3-27b-pt`]() using mlx-vlm version **0.1.17**.
 Refer to the [original model card](https://huggingface.co/google/gemma-3-27b-pt) for more details on the model.
 ## Use with mlx

 ---
 # mlx-community/gemma-3-27b-pt-8bit
+This model was converted to MLX format from [`google/gemma-3-27b-pt`]() using mlx-vlm version **0.1.18**.
 Refer to the [original model card](https://huggingface.co/google/gemma-3-27b-pt) for more details on the model.
 ## Use with mlx

+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.50.0.dev0"
+}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5faddeaf9ed9651c4fd507a4de9c2eced6a8712d082639365d867ad0689d265d
-size 5279749711

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97a038eb3fc2ab0879326dee9c00cea34a4a051167b24f14092ab6eaf47e6102
-size 5264743290

 version https://git-lfs.github.com/spec/v1
+oid sha256:06423ed600bd8d7365d3f67a5b264976083cebb96c72e5f68f6cd704a641559d
+size 5279750284

 version https://git-lfs.github.com/spec/v1
+oid sha256:60b5120e18fcb6044b798ae283f3a00bd6b56f214b9842f84ec597072c25fa33
+size 5264743530

 version https://git-lfs.github.com/spec/v1
+oid sha256:f947aa32476a21e0f472e4247b14144d48386a7edc02805d8d7af92419de7dfa
+size 5264743642

 version https://git-lfs.github.com/spec/v1
+oid sha256:a019b479c3aea8b4e316a41570fc649a691f52351c7cf26db4e2bfb8efb2227d
+size 5264743640

 version https://git-lfs.github.com/spec/v1
+oid sha256:c71103fc438d1369c59853ef2cd9cd3f902b76f7cd46bce6f19f21d7ea85428a
+size 5264743646

 version https://git-lfs.github.com/spec/v1
+oid sha256:5aebaa611f57538fdcf6036f8882215bfdb8315f35b2c1af1feb152fe4f072bd
+size 4704107440

+{
+  "image_seq_length": 256,
+  "processor_class": "Gemma3Processor"
+}

	@@ -0,0 +1,8 @@

	@@ -0,0 +1,4 @@