fix ltx model selection

Signed-off-by: Vladimir Mandic <mandic00@live.com>

CHANGELOG.md

@@ -108,9 +108,10 @@
   - reference: fix download for sd15/sdxl reference models  
   - fix checks in init/mask image decode  
   - fix hf token with extra chars  
-  - **FramePack** job state tracking  
-  - **LTXVideo** with offloading  
-  - **WAN** with offloading  
+  - **FramePack** add job state tracking  
+  - **LTXVideo** fix model selection in ltx tab
+  - **LTXVideo** fix run with offloading  
+  - **WAN** fix run with offloading  
 
 ## Update for 2025-09-15
 

modules/ltx/ltx_process.py

@@ -13,7 +13,7 @@ from modules.ltx.ltx_util import get_bucket, get_frames, load_model, load_upsamp
 
 
 debug = shared.log.trace if os.environ.get('SD_VIDEO_DEBUG', None) is not None else lambda *args, **kwargs: None
-engine, model = 'LTX Video', 'LTXVideo 0.9.7 13B'
+# engine, model = 'LTX Video', 'LTXVideo 0.9.7 13B'
 upsample_repo_id = "a-r-r-o-w/LTX-Video-0.9.7-Latent-Spatial-Upsampler-diffusers"
 upsample_pipe = None
 queue_lock = threading.Lock()
@@ -21,6 +21,7 @@ queue_lock = threading.Lock()
 
 def run_ltx(task_id,
             _ui_state,
+            model:str,
             prompt:str,
             negative:str,
             styles:list[str],
@@ -73,6 +74,7 @@ def run_ltx(task_id,
         memstats.reset_stats()
         timer.process.reset()
         yield None, 'LTX: Loading...'
+        engine = 'LTX Video'
         load_model(engine, model)
 
         shared.state.begin('Video', task_id=task_id)

modules/ltx/ltx_ui.py

@@ -10,18 +10,6 @@ from modules.ltx import ltx_process
 debug = shared.log.trace if os.environ.get('SD_VIDEO_DEBUG', None) is not None else lambda *args, **kwargs: None
 
 
-def load_model(model_name):
-    if model_name is None or model_name == 'None':
-        shared.log.info('LTX model unload')
-        from modules import sd_models
-        sd_models.unload_model_weights()
-        return
-    else:
-        model_instance = [m for m in models['LTX Video'] if m.name == model_name][0]
-        from modules.video_models import video_load
-        video_load.load_model(model_instance)
-
-
 def create_ui(prompt, negative, styles, overrides):
     with gr.Row():
         with gr.Column(variant='compact', elem_id="ltx_settings", elem_classes=['settings-column'], scale=1):
@@ -30,7 +18,6 @@ def create_ui(prompt, negative, styles, overrides):
             with gr.Row():
                 ltx_models = [m.name for m in models['LTX Video']]
                 model = gr.Dropdown(label='LTX model', choices=ltx_models, value=ltx_models[0])
-                model.change(fn=load_model, inputs=[model], outputs=[], show_progress=True)
             with gr.Accordion(open=True, label="LTX size", elem_id='ltx_generate_accordion'):
                 with gr.Row():
                     width, height = ui_sections.create_resolution_inputs('ltx', default_width=832, default_height=480)
@@ -90,6 +77,7 @@ def create_ui(prompt, negative, styles, overrides):
     state_inputs = [task_id, ui_state]
 
     video_inputs = [
+        model,
         prompt, negative, styles,
         width, height, frames,
         steps, sampler_index, seed,

modules/ltx/ltx_util.py

@@ -4,6 +4,9 @@ from PIL import Image
 from modules import devices, shared, sd_models, timer, extra_networks
 
 
+loaded_model: str = None
+
+
 def get_bucket(size: int):
     return int(size) - (int(size) % shared.sd_model.vae_temporal_compression_ratio)
 
@@ -13,11 +16,15 @@ def get_frames(frames: int):
 
 
 def load_model(engine: str, model: str):
+    global loaded_model # pylint: disable=global-statement
+    if loaded_model == model:
+        return
     t0 = time.time()
     from modules.video_models import models_def, video_load
     selected: models_def.Model = [m for m in models_def.models[engine] if m.name == model][0]
-    shared.log.info(f'Video load: cls={selected.repo_cls.__name__} repo="{selected.repo}"')
+    shared.log.info(f'Video load: engine="{engine}" selected="{model}" {selected}')
     video_load.load_model(selected)
+    loaded_model = model
     t1 = time.time()
     shared.sd_model = sd_models.apply_balanced_offload(shared.sd_model)
     t2 = time.time()

modules/video_models/video_ui.py

@@ -1,6 +1,6 @@
 import os
 import gradio as gr
-from modules import shared, sd_models, ui_common, ui_sections, ui_symbols, call_queue
+from modules import shared, sd_models, ui_common, ui_sections, ui_symbols, ui_video_vlm, call_queue
 from modules.ui_components import ToolButton
 from modules.video_models import models_def, video_utils
 from modules.video_models import video_run
@@ -112,6 +112,7 @@ def create_ui(prompt, negative, styles, overrides):
                 init_image = gr.Image(elem_id="video_image", show_label=False, type="pil", image_mode="RGB", width=256, height=256)
                 gr.HTML("<br>&nbsp Last image")
                 last_image = gr.Image(elem_id="video_last", show_label=False, type="pil", image_mode="RGB", width=256, height=256)
+            vlm_enhance, vlm_model, vlm_system_prompt = ui_video_vlm.create_ui(prompt_element=prompt, image_element=init_image)
             with gr.Accordion(open=False, label="Output", elem_id='video_output_accordion'):
                 with gr.Row():
                     save_frames = gr.Checkbox(label='Save image frames', value=False, elem_id="video_save_frames")