Retry long videos decode with tiling fallback

2026-08-30 20:28:47 +00:00
parent dfead66b9e
commit 6a7bec5d03
2 changed files with 114 additions and 11 deletions
@@ -6372,16 +6372,29 @@ class H3LongVideos:
                    e._h3_stage = "sampling"
                raise
            refined_out = _video_only_refined_latent(out, refined_out)
-        # Keep a CPU copy of the sampled latent BEFORE decoding, for the `latent`
-        # output. Latents are ~1000x smaller than the frames they decode to (a
-        # 1344x768 124f shot is ~1.5MB against ~1.5GB), so carrying one per shot for
-        # the whole chain is free. Detached and moved off the card immediately, for
-        # the same reason the decoded frames are.
-        decode_video_start = time.perf_counter()
-        shot_latent = _copy_sample_latent(refined_out)
-        video = _decode_video(vae, refined_out, tiled, free_first=model,
-                              tile_t=decode_tile_frames, tile_xy=decode_tile_size)
-        timing["decode_video"] += time.perf_counter() - decode_video_start
+        # Keep a CPU copy of the sampled latent BEFORE decoding, for the `latent`
+        # output. Latents are ~1000x smaller than the frames they decode to (a
+        # 1344x768 124f shot is ~1.5MB against ~1.5GB), so carrying one per shot for
+        # the whole chain is free. Detached and moved off the card immediately, for
+        # the same reason the decoded frames are.
+        decode_video_start = time.perf_counter()
+        shot_latent = _copy_sample_latent(refined_out)
+        try:
+            video = _decode_video(vae, refined_out, tiled, free_first=model,
+                                  tile_t=decode_tile_frames, tile_xy=decode_tile_size)
+        except Exception as e:
+            # Decode is the biggest allocation in the run. If the straight path
+            # fails, retry once with tiled decode and without the aggressive unload
+            # so a marginal card can still finish the render.
+            if not _is_oom(e) and "decode" not in str(e).lower():
+                raise
+            mm.soft_empty_cache(True)
+            retry_tiled = True
+            retry_tile_t = decode_tile_frames or 16
+            retry_tile_xy = decode_tile_size or 256
+            video = _decode_video(vae, refined_out, retry_tiled, free_first=None,
+                                  tile_t=retry_tile_t, tile_xy=retry_tile_xy)
+        timing["decode_video"] += time.perf_counter() - decode_video_start
        decode_audio_start = time.perf_counter()
        audio = _decode_audio(audio_vae, out)
        timing["decode_audio"] += time.perf_counter() - decode_audio_start