Add GPU (NVENC AV1) webm format and save_latent toggle to FastAbsoluteSaver

- nvenc_av1-webm: GPU-encoded webm via av1_nvenc (NVENC has no VP9
  encoder); uses libopus since webm rejects AAC audio.
- save_latent boolean gates the latent sidecar write while keeping the
  latent passthrough intact.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

fast_saver.py

@@ -139,6 +139,10 @@ VIDEO_FORMATS = {
     "nvenc_av1-mp4": {"ext": ".mp4",  "codec": ["-c:v", "av1_nvenc"],
                       "quality": "bitrate", "color_mgmt": True, "acodec": "aac",
                       "extra": ["-movflags", "+faststart"]},
+    # GPU webm: NVENC has no VP9 encoder, so use AV1 (a valid WebM codec) on the GPU.
+    # WebM containers only allow Opus/Vorbis audio, not AAC.
+    "nvenc_av1-webm":{"ext": ".webm", "codec": ["-c:v", "av1_nvenc"],
+                      "quality": "bitrate", "color_mgmt": True, "acodec": "libopus"},
 }
 
 
@@ -164,7 +168,8 @@ class FastAbsoluteSaver:
 
                 # --- FORMAT SWITCH ---
                 "save_format": (["png", "webp", "mp4", "webm", "h265-mp4", "av1-mp4", "gif",
-                                 "ffv1-mkv", "prores-mov", "nvenc_h264-mp4", "nvenc_hevc-mp4", "nvenc_av1-mp4"], ),
+                                 "ffv1-mkv", "prores-mov", "nvenc_h264-mp4", "nvenc_hevc-mp4", "nvenc_av1-mp4",
+                                 "nvenc_av1-webm"], ),
 
                 # --- NAMING CONTROL ---
                 "use_timestamp": ("BOOLEAN", {"default": False, "label": "Add Timestamp (Unique)"}),
@@ -176,6 +181,7 @@ class FastAbsoluteSaver:
                 "metadata_key": ("STRING", {"default": "sharpness_score"}),
                 "save_workflow_metadata": ("BOOLEAN", {"default": False, "label": "Save ComfyUI Workflow (Graph)"}),
                 "save_metadata_png": ("BOOLEAN", {"default": False, "label": "Embed Workflow in PNG (sidecar or first file)"}),
+                "save_latent": ("BOOLEAN", {"default": True, "label": "Save Latent Sidecar"}),
 
                 # --- PERFORMANCE ---
                 "max_threads": ("INT", {"default": 0, "min": 0, "max": 128, "step": 1, "label": "Max Threads (0=Auto)"}),
@@ -504,7 +510,7 @@ class FastAbsoluteSaver:
         return out_file
 
     def save_images_fast(self, images, output_path, filename_prefix, save_format, use_timestamp, auto_increment, counter_digits,
-                         max_threads, filename_with_score, metadata_key, save_workflow_metadata, save_metadata_png,
+                         max_threads, filename_with_score, metadata_key, save_workflow_metadata, save_metadata_png, save_latent,
                          webp_lossless, webp_quality, webp_method,
                          video_fps, video_crf, video_pixel_format,
                          video_bitrate, prores_profile, gif_dither,
@@ -538,7 +544,7 @@ class FastAbsoluteSaver:
                             extra_data=extra_pnginfo,
                             bitrate=video_bitrate, prores_profile=prores_profile,
                             gif_dither=gif_dither, audio=audio)
-            if latent is not None:
+            if latent is not None and save_latent:
                 self._save_latent_sidecar(latent, out_file)
             # Save metadata sidecar PNG next to the video file
             if save_metadata_png:
@@ -612,7 +618,7 @@ class FastAbsoluteSaver:
                 if future.result():
                     saved_image_paths.append(full_path)
 
-        if latent is not None:
+        if latent is not None and save_latent:
             for image_path in saved_image_paths:
                 self._save_latent_sidecar(latent, image_path)
 

tests/test_fast_saver_latent.py

@@ -4,7 +4,7 @@ import torch
 from fast_saver import FastAbsoluteSaver
 
 
-def _save_args(tmp_path, *, save_format="png", latent=None):
+def _save_args(tmp_path, *, save_format="png", latent=None, save_latent=True):
     return {
         "images": torch.zeros((1, 2, 2, 3), dtype=torch.float32),
         "output_path": str(tmp_path),
@@ -18,6 +18,7 @@ def _save_args(tmp_path, *, save_format="png", latent=None):
         "metadata_key": "sharpness_score",
         "save_workflow_metadata": False,
         "save_metadata_png": False,
+        "save_latent": save_latent,
         "webp_lossless": True,
         "webp_quality": 100,
         "webp_method": 4,
@@ -55,6 +56,17 @@ def test_png_save_returns_latent_passthrough(tmp_path):
     assert result["result"][0] is latent
 
 
+def test_png_save_latent_false_skips_sidecar_but_keeps_passthrough(tmp_path):
+    saver = FastAbsoluteSaver()
+    latent = {"samples": torch.ones((1, 1, 2, 2))}
+
+    result = saver.save_images_fast(**_save_args(tmp_path, latent=latent, save_latent=False))
+
+    assert not (tmp_path / "frame_0000.latent").exists()
+    assert result["result"] == (latent,)
+    assert result["result"][0] is latent
+
+
 def test_video_save_writes_latent_sidecar_next_to_video(tmp_path):
     saver = FastAbsoluteSaver()
     latent = {"samples": torch.arange(8, dtype=torch.float32).reshape(2, 1, 2, 2)}