Instructions to use pytorch/gemma-3-12b-it-INT4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pytorch/gemma-3-12b-it-INT4 with Transformers:

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

pipe = pipeline("image-text-to-text", model="pytorch/gemma-3-12b-it-INT4")
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)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("pytorch/gemma-3-12b-it-INT4")
model = AutoModelForMultimodalLM.from_pretrained("pytorch/gemma-3-12b-it-INT4", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use pytorch/gemma-3-12b-it-INT4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "pytorch/gemma-3-12b-it-INT4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pytorch/gemma-3-12b-it-INT4",
		"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/pytorch/gemma-3-12b-it-INT4

SGLang

How to use pytorch/gemma-3-12b-it-INT4 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 "pytorch/gemma-3-12b-it-INT4" \
    --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": "pytorch/gemma-3-12b-it-INT4",
		"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 "pytorch/gemma-3-12b-it-INT4" \
        --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": "pytorch/gemma-3-12b-it-INT4",
		"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"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use pytorch/gemma-3-12b-it-INT4 with Docker Model Runner:
```
docker model run hf.co/pytorch/gemma-3-12b-it-INT4
```

appy1234 commited on Mar 11

Commit

b53ff8e

verified ·

1 Parent(s): 1dd1975

Upload Gemma3ForConditionalGeneration

Browse files

Files changed (4) hide show

config.json +6 -16
generation_config.json +1 -1
pytorch_model-00001-of-00002.bin +1 -1
pytorch_model-00002-of-00002.bin +1 -1

config.json CHANGED Viewed

@@ -37,9 +37,7 @@
     "attention_bias": false,
     "attention_dropout": 0.0,
     "attn_logit_softcapping": null,
-    "bos_token_id": 2,
     "dtype": "bfloat16",
-    "eos_token_id": 1,
     "final_logit_softcapping": null,
     "head_dim": 256,
     "hidden_activation": "gelu_pytorch_tanh",
@@ -101,28 +99,20 @@
     "num_attention_heads": 16,
     "num_hidden_layers": 48,
     "num_key_value_heads": 8,
-    "pad_token_id": 0,
     "query_pre_attn_scalar": 256,
     "rms_norm_eps": 1e-06,
-    "rope_parameters": {
-      "full_attention": {
-        "factor": 8.0,
-        "rope_theta": 1000000.0,
-        "rope_type": "linear"
-      },
-      "sliding_attention": {
-        "rope_theta": 10000.0,
-        "rope_type": "default"
-      }
     },
     "sliding_window": 1024,
-    "tie_word_embeddings": true,
     "use_bidirectional_attention": false,
     "use_cache": true,
     "vocab_size": 262208
   },
-  "tie_word_embeddings": true,
-  "transformers_version": "5.3.0.dev0",
   "vision_config": {
     "attention_dropout": 0.0,
     "dtype": "bfloat16",

     "attention_bias": false,
     "attention_dropout": 0.0,
     "attn_logit_softcapping": null,
     "dtype": "bfloat16",
     "final_logit_softcapping": null,
     "head_dim": 256,
     "hidden_activation": "gelu_pytorch_tanh",
     "num_attention_heads": 16,
     "num_hidden_layers": 48,
     "num_key_value_heads": 8,
     "query_pre_attn_scalar": 256,
     "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_bidirectional_attention": false,
     "use_cache": true,
     "vocab_size": 262208
   },
+  "transformers_version": "4.57.5",
   "vision_config": {
     "attention_dropout": 0.0,
     "dtype": "bfloat16",

generation_config.json CHANGED Viewed

@@ -9,5 +9,5 @@
   "pad_token_id": 0,
   "top_k": 64,
   "top_p": 0.95,
-  "transformers_version": "5.3.0.dev0"
 }

   "pad_token_id": 0,
   "top_k": 64,
   "top_p": 0.95,
+  "transformers_version": "4.57.5"
 }

pytorch_model-00001-of-00002.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9aab462567b1a65a7b9e098bd51a3b07c8b1adfd9a43c53d516098ad6b81adcf
 size 4987220254

 version https://git-lfs.github.com/spec/v1
+oid sha256:63a7663d62a5836930dafbaafbff8677a8ede00d7fea25f02edae47be7582af7
 size 4987220254

pytorch_model-00002-of-00002.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d56c43211b1881c675d3be98f3bfda1bc19b5f9f8f3b8b2c5ff610697927203
 size 3548065976

 version https://git-lfs.github.com/spec/v1
+oid sha256:5b4267d1d5a99e0fb1ee13b147870388144f86244bf5b72ddd350298bee63e73
 size 3548065976