[None][feat] Add LTX-2 visual generation example

examples/visual_gen/README.md

@@ -21,6 +21,7 @@ python models/wan_t2v.py
 
 # With engine config (quant, parallelism, etc.)
 python models/wan_t2v.py --visual_gen_args configs/wan2.2-t2v-fp4-1gpu.yaml
+python models/ltx2.py --visual_gen_args configs/ltx2.yaml
 ```
 
 Install deps from the repo root: `pip install -r requirements-dev.txt`.

examples/visual_gen/configs/ltx2.yaml

@@ -0,0 +1,33 @@
+# SPDX-FileCopyrightText: Copyright (c) 2022-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# 1-GPU LTX-2 text-to-video with audio.
+# Shared by offline examples (--visual_gen_args) and trtllm-serve.
+#
+# LTX-2 stores the Gemma text encoder separately from the diffusion checkpoint.
+# Set text_encoder_path to the local Gemma3 checkpoint directory before running
+# against a local LTX-2 checkpoint.
+pipeline_config:
+  text_encoder_path: google/gemma-3-12b-it
+
+attention_config:
+  backend: VANILLA
+
+parallel_config:
+  cfg_size: 1
+  ulysses_size: 1
+
+cuda_graph_config:
+  enable: false

examples/visual_gen/models/ltx2.py

@@ -0,0 +1,95 @@
+#!/usr/bin/env python3
+# SPDX-FileCopyrightText: Copyright (c) 2022-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""LTX-2 Text-to-Video generation with audio.
+
+Usage:
+    python ltx2.py
+    python ltx2.py --visual_gen_args ../configs/ltx2.yaml
+"""
+
+import argparse
+
+from tensorrt_llm import VisualGen, VisualGenArgs
+
+
+def main():
+    parser = argparse.ArgumentParser(description="LTX-2 Text-to-Video example")
+    parser.add_argument(
+        "--model",
+        type=str,
+        default="Lightricks/LTX-2",
+        help="Model path or HuggingFace Hub ID",
+    )
+    parser.add_argument(
+        "--visual_gen_args",
+        "--extra_visual_gen_options",
+        dest="visual_gen_args",
+        type=str,
+        default=None,
+        help="Path to YAML config (same as trtllm-serve --visual_gen_args)",
+    )
+    parser.add_argument(
+        "--text_encoder_path",
+        type=str,
+        default=None,
+        help=(
+            "Gemma3 text encoder path. Overrides pipeline_config.text_encoder_path "
+            "from --visual_gen_args when set."
+        ),
+    )
+    parser.add_argument(
+        "--output_path",
+        type=str,
+        default="ltx2_t2v_output.mp4",
+        help="Path to save the output video",
+    )
+    args = parser.parse_args()
+
+    # LTX-2 requires pipeline_config.text_encoder_path for the Gemma3 text
+    # encoder. The YAML path is preferred for production configs; the default
+    # below keeps this script runnable as a minimal offline example.
+    extra_args = VisualGenArgs.from_yaml(args.visual_gen_args) if args.visual_gen_args else VisualGenArgs()
+    text_encoder_path = args.text_encoder_path
+    if text_encoder_path is None and not args.visual_gen_args:
+        text_encoder_path = "google/gemma-3-12b-it"
+    if text_encoder_path is not None:
+        extra_args.pipeline_config = {
+            **extra_args.pipeline_config,
+            "text_encoder_path": text_encoder_path,
+        }
+    visual_gen = VisualGen(model=args.model, args=extra_args)
+
+    # --- Model-specific: T2V request construction ---
+    # Start from LTX-2 defaults and override the main request shape explicitly.
+    params = visual_gen.default_params
+    params.height = 512
+    params.width = 768
+    params.num_frames = 121
+    params.frame_rate = 24.0
+    params.num_inference_steps = 40
+    params.guidance_scale = 4.0
+
+    output = visual_gen.generate(
+        inputs="A cinematic shot of a cat walking through a field of flowers",
+        params=params,
+    )
+
+    output.save(args.output_path)
+    print(f"Saved: {args.output_path}")
+
+
+if __name__ == "__main__":
+    main()