You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 10424ee
Browse filesBrowse the repository at this point in the historyBrowse files
Copy file name to clipboardExpand all lines: README.md
+10-8Lines changed: 10 additions & 8 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -94,14 +94,16 @@ Author-reported means from three requests after two warmups; not independently v
94
94
95
95
Author-reported results on **Apple M5 Max (40-core GPU), 128 GB, macOS 27**, at **768×448, 49 frames, seed 0**, using real weights with one engine running at a time.
96
96
97
-
| Engine | Steps | Cold latency (s) ↓ | Warm latency (s) ↓ |
Cold includes the first clip's weight loading, text encoding and preparation; warm reuses loaded weights and prepared state. FastH3 uses a distilled 4-step checkpoint, unlike the 20-step configurations. Warm values average clips 2–4 for the 20-step rows and two A/B/B/A warm runs for FastH3.
Cold includes the first clip's weight loading, text encoding and preparation; warm reuses loaded weights and prepared state.
105
+
106
+
¹ FastH3 uses a distilled 4-step checkpoint, unlike the 20-step configurations. Warm values average clips 2–4 for the 20-step rows and two A/B/B/A warm runs for FastH3.
105
107
106
108
[Author's report and reproduction →](https://github.com/zuiho-kai/h3-metal/blob/53b7fa5d88e2497ab427ab9c7155535f7e6e6817/README.md#performance). These external results were not rerun in InferRSI.
0 commit comments