diff --git a/CHANGELOG.md b/CHANGELOG.md
index 12d233b00..d1e009550 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -8,6 +8,8 @@
     see [Nunchaku Wiki](https://github.com/vladmandic/sdnext/wiki/Nunchaku) for installation guide and list of supported models & features  
   - [LTXVideo 0.9.6](https://github.com/Lightricks/LTX-Video?tab=readme-ov-file) T2V and I2V  
     in both **Standard** and **Distilled** variants  
+  - [WAN 2.1 FLF2V 14B 720P](https://huggingface.co/Wan-AI/Wan2.1-FLF2V-14B-720P)  
+    new first-to-last image video model from WAN-AI  
   - [CFG-Zero](https://github.com/WeichenFan/CFG-Zero-star) new guidance method optimized for flow-matching models  
     implemented for **FLUX.1, HiDream-I1, SD3.x, CogView4, HunyuanVideo, WanAI**  
     enable and configure in *settings -> pipeline modifiers -> cfg zero*  
diff --git a/modules/ui_video.py b/modules/ui_video.py
index b53810d4b..2fd23778c 100644
--- a/modules/ui_video.py
+++ b/modules/ui_video.py
@@ -114,9 +114,11 @@ def create_ui():
                         vae_type = gr.Dropdown(label='VAE decode', choices=['Default', 'Tiny', 'Remote'], value='Default', elem_id="video_vae_type")
                         vae_tile_frames = gr.Slider(label='Tile frames', minimum=1, maximum=64, step=1, value=16, elem_id="video_vae_tile_frames")
                 with gr.Accordion(open=False, label="Init image", elem_id='video_init_accordion'):
+                    init_strength = gr.Slider(label='Init strength', minimum=0.0, maximum=1.0, step=0.01, value=0.5, elem_id="video_denoising_strength")
                     gr.HTML("<br>&nbsp Init image")
                     init_image = gr.Image(elem_id="video_image", show_label=False, type="pil", image_mode="RGB", height=512)
-                    init_strength = gr.Slider(label='Init strength', minimum=0.0, maximum=1.0, step=0.01, value=0.5, elem_id="video_denoising_strength")
+                    gr.HTML("<br>&nbsp Last image")
+                    last_image = gr.Image(elem_id="video_last", show_label=False, type="pil", image_mode="RGB", height=512)
                 with gr.Accordion(open=True, 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")
@@ -167,7 +169,7 @@ def create_ui():
             sampler_shift, dynamic_shift,
             seed,
             guidance_scale, guidance_true,
-            init_image, init_strength,
+            init_image, init_strength, last_image,
             vae_type, vae_tile_frames,
             save_frames,
             video_type, video_duration, video_loop, video_pad, video_interpolate,
diff --git a/modules/video_models/models_def.py b/modules/video_models/models_def.py
index 0d471cd60..f3c69c756 100644
--- a/modules/video_models/models_def.py
+++ b/modules/video_models/models_def.py
@@ -158,6 +158,12 @@ models = {
               repo_cls=diffusers.WanImageToVideoPipeline,
               te_cls=transformers.T5EncoderModel,
               dit_cls=diffusers.WanTransformer3DModel),
+        Model(name='WAN 2.1 14B FLF2V 720p',
+              url='https://huggingface.co/Wan-AI/Wan2.1-FLF2V-14B-720P',
+              repo='Wan-AI/Wan2.1-FLF2V-14B-720P-diffusers',
+              repo_cls=diffusers.WanImageToVideoPipeline,
+              te_cls=transformers.T5EncoderModel,
+              dit_cls=diffusers.WanTransformer3DModel),
     ],
     'Mochi Video': [
         Model(name='None'),
diff --git a/modules/video_models/video_overrides.py b/modules/video_models/video_overrides.py
index 4168a1441..655256e95 100644
--- a/modules/video_models/video_overrides.py
+++ b/modules/video_models/video_overrides.py
@@ -14,7 +14,7 @@ def load_override(selected: Model):
         kwargs['vae'] = diffusers.AutoencoderKLAllegro.from_pretrained(selected.repo, subfolder="vae", torch_dtype=torch.float32, cache_dir=shared.opts.hfcache_dir)
     if selected.name == 'LTXVideo 0.9.5 I2V':
         kwargs['vae'] = diffusers.AutoencoderKLLTXVideo.from_pretrained(selected.repo, subfolder="vae", torch_dtype=torch.float32, cache_dir=shared.opts.hfcache_dir)
-    if selected.name == 'WAN 2.1 14B I2V 480p' or selected.name == 'WAN 2.1 14B I2V 720p':
+    if 'WAN 2.1 14B' in selected.name:
         kwargs['vae'] = diffusers.AutoencoderKLWan.from_pretrained(selected.repo, subfolder="vae", torch_dtype=torch.float32, cache_dir=shared.opts.hfcache_dir)
     debug(f'Video overrides: model="{selected.name}" kwargs={list(kwargs)}')
     return kwargs
diff --git a/modules/video_models/video_run.py b/modules/video_models/video_run.py
index b0248e914..6a3e13ebb 100644
--- a/modules/video_models/video_run.py
+++ b/modules/video_models/video_run.py
@@ -8,7 +8,7 @@ debug = shared.log.trace if os.environ.get('SD_VIDEO_DEBUG', None) is not None e
 
 
 def generate(*args, **kwargs):
-    task_id, ui_state, engine, model, prompt, negative, styles, width, height, frames, steps, sampler_index, sampler_shift, dynamic_shift, seed, guidance_scale, guidance_true, init_image, init_strength, vae_type, vae_tile_frames, save_frames, video_type, video_duration, video_loop, video_pad, video_interpolate, override_settings = args
+    task_id, ui_state, engine, model, prompt, negative, styles, width, height, frames, steps, sampler_index, sampler_shift, dynamic_shift, seed, guidance_scale, guidance_true, init_image, init_strength, last_image, vae_type, vae_tile_frames, save_frames, video_type, video_duration, video_loop, video_pad, video_interpolate, override_settings = args
     if engine is None or model is None or engine == 'None' or model == 'None':
         return video_utils.queue_err('model not selected')
     found = [model.name for model in models_def.models.get(engine, [])]
@@ -57,9 +57,17 @@ def generate(*args, **kwargs):
             return video_utils.queue_err('init image not set')
         p.task_args['image'] = images.resize_image(resize_mode=2, im=init_image, width=p.width, height=p.height, upscaler_name=None, output_type='pil')
         shared.log.debug(f'Video: op=I2V init={init_image} resized={p.task_args["image"]}')
+    elif 'FLF2V' in model:
+        if init_image is None:
+            return video_utils.queue_err('init image not set')
+        if last_image is None:
+            return video_utils.queue_err('last image not set')
+        p.task_args['image'] = images.resize_image(resize_mode=2, im=init_image, width=p.width, height=p.height, upscaler_name=None, output_type='pil')
+        p.task_args['last_image'] = images.resize_image(resize_mode=2, im=last_image, width=p.width, height=p.height, upscaler_name=None, output_type='pil')
+        shared.log.debug(f'Video: op=FLF2V init={init_image} last={last_image} resized={p.task_args["image"]}')
     elif 'T2V' in model:
         if init_image is not None:
-            shared.log.debug('Video: op=T2V init image not supported')
+            shared.log.warning('Video: op=T2V init image not supported')
 
     # cleanup memory
     shared.sd_model = sd_models.apply_balanced_offload(shared.sd_model)