[Tracing] Support tracing of Gemma3 [#1248] (#1373)

SUMMARY:
Add support for tracing of Gemma3:
[issue#1248](#1248).

Steps that I have done:
1. Create gemma3.py from HF and update __init__.py.
2. Classes and functions that I modified:
    2.1 Gemma3ForConditionalGeneration: _update_causal_mask and forward
2.2 Gemma3TextModel: _update_causal_mask, forward, and
_prepare_4d_causal_attention_mask_with_cache_position


TEST PLAN:
Ran:
`llmcompressor.trace --model_id google/gemma-3-4b-it --model_class
TraceableGemma3ForConditionalGeneration --ignore "lm_head"
"re:vision_tower.*" --modality vision`

Output:
<img width="796" alt="trace_output"
src="https://github.com/user-attachments/assets/8f5c9c7d-32a9-4b12-b4b2-10b6a4352846"
/>

This is my first attempt at solving this issue. It is a fun learning
experience and please review it carefully.
Gemma3 can go through tracing now, but we might need further tests for
the quantization as well.

---------

Signed-off-by: Kelvin Cheng <kelvincheng216@gmail.com>
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
Signed-off-by: Rahul Tuli <rtuli@redhat.com>
Signed-off-by: Brian Dellabetta <bdellabe@redhat.com>
Signed-off-by: Domenic Barbuzzi <dbarbuzz@redhat.com>
Co-authored-by: Kyle Sayers <kylesayrs@gmail.com>
Co-authored-by: Dipika Sikka <dipikasikka1@gmail.com>
Co-authored-by: Vedant <146507396+ved1beta@users.noreply.github.com>
Co-authored-by: Rahul Tuli <rtuli@redhat.com>
Co-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>
Co-authored-by: Domenic Barbuzzi <dbarbuzz@redhat.com>

Author: 7 people

examples/multimodal_vision/gemma3_example.py

@@ -0,0 +1,75 @@
+import requests
+import torch
+from PIL import Image
+from transformers import AutoProcessor
+
+from llmcompressor import oneshot
+from llmcompressor.modifiers.quantization import GPTQModifier
+from llmcompressor.transformers.tracing import TraceableGemma3ForConditionalGeneration
+
+# Load model.
+model_id = "google/gemma-3-4b-it"
+model = TraceableGemma3ForConditionalGeneration.from_pretrained(
+    model_id, device_map="auto", torch_dtype="auto"
+)
+processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
+
+# Oneshot arguments
+DATASET_ID = "flickr30k"
+DATASET_SPLIT = {"calibration": "test[:512]"}
+NUM_CALIBRATION_SAMPLES = 512
+MAX_SEQUENCE_LENGTH = 2048
+
+
+# Define a oneshot data collator for multimodal inputs.
+def data_collator(batch):
+    assert len(batch) == 1
+    return {key: torch.tensor(value) for key, value in batch[0].items()}
+
+
+# Recipe
+recipe = [
+    GPTQModifier(
+        targets="Linear",
+        scheme="W4A16",
+        ignore=["re:*.lm_head", "re:vision_tower.*", "re:multi_modal_projector.*"],
+    ),
+]
+
+# Perform oneshot
+oneshot(
+    model=model,
+    tokenizer=model_id,
+    dataset=DATASET_ID,
+    splits=DATASET_SPLIT,
+    recipe=recipe,
+    max_seq_length=MAX_SEQUENCE_LENGTH,
+    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
+    trust_remote_code_model=True,
+    data_collator=data_collator,
+)
+
+# Confirm generations of the quantized model look sane.
+print("========== SAMPLE GENERATION ==============")
+messages = [
+    {
+        "role": "user",
+        "content": [
+            {"type": "text", "text": "Please describe the animal in this image\n"},
+            {"type": "image"},
+        ],
+    },
+]
+prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
+image_url = "http://images.cocodataset.org/train2017/000000231895.jpg"
+raw_image = Image.open(requests.get(image_url, stream=True).raw)
+
+inputs = processor(images=raw_image, text=prompt, return_tensors="pt").to("cuda")
+output = model.generate(**inputs, max_new_tokens=100)
+print(processor.decode(output[0], skip_special_tokens=True))
+print("==========================================")
+
+# Save to disk compressed.
+SAVE_DIR = model_id.split("/")[1] + "-W4A16-G128"
+model.save_pretrained(SAVE_DIR, save_compressed=True)
+processor.save_pretrained(SAVE_DIR)

src/llmcompressor/transformers/tracing/__init__.py

@@ -1,3 +1,6 @@
+from .gemma3 import (
+    Gemma3ForConditionalGeneration as TraceableGemma3ForConditionalGeneration,
+)
 from .llava import (
     LlavaForConditionalGeneration as TraceableLlavaForConditionalGeneration,
 )
@@ -11,12 +14,13 @@
     Idefics3ForConditionalGeneration as TraceableIdefics3ForConditionalGeneration,
 )
 from .qwen2_5_vl import (
-    Qwen2_5_VLForConditionalGeneration as TraceableQwen2_5_VLForConditionalGeneration
+    Qwen2_5_VLForConditionalGeneration as TraceableQwen2_5_VLForConditionalGeneration,
 )
 from .debug import get_model_class
 
 __all__ = [
     "get_model_class",
+    "TraceableGemma3ForConditionalGeneration",
     "TraceableLlavaForConditionalGeneration",
     "TraceableMllamaForConditionalGeneration",
     "TraceableQwen2VLForConditionalGeneration",