mlboydaisuke commited on
Commit
87bccbb
·
verified ·
1 Parent(s): 5ec4060

Add measured Raspberry Pi 5 (CPU) section

Browse files

Numbers from the 2026-08-31 Raspberry Pi 5 sweep: LiteRT benchmark_model
(litert-cli-nightly 0.2.0.dev20260805), CPU/XNNPACK, 4 threads, 10 warm-up +
50 timed runs x 3 invocations per file, cooldown between invocations, thermal
throttle state verified 0x0 on every run. Median + spread, never single-shot.
Measurements are joined to this repo by exact .tflite file name.

Files changed (1) hide show
  1. README.md +8 -0
README.md CHANGED
@@ -114,6 +114,14 @@ Measured on a **Samsung Galaxy S26** (Snapdragon 8 Elite Gen 5 / SM8850, Hexagon
114
 
115
  GPU wiring: [GPU guide](https://github.com/john-rocky/hf-to-litertlm/blob/main/docs/android-gpu.md).
116
 
 
 
 
 
 
 
 
 
117
  ## License
118
 
119
  [MIT](https://github.com/megvii-research/NAFNet/blob/main/LICENSE). Upstream:
 
114
 
115
  GPU wiring: [GPU guide](https://github.com/john-rocky/hf-to-litertlm/blob/main/docs/android-gpu.md).
116
 
117
+ ## Raspberry Pi 5 (CPU)
118
+
119
+ Measured on a Raspberry Pi 5 Model B Rev 1.1 (8 GB, Raspberry Pi OS 64-bit) with the LiteRT [`benchmark_model`](https://ai.google.dev/edge/litert/models/measurement) tool from litert-cli-nightly 0.2.0.dev20260805: CPU inference (XNNPACK, 4 threads), 3 invocations per file of 10 warm-up plus 50 timed runs (the tool caps a phase at 150 s, so very slow graphs run fewer — the Runs column is the actual timed total). The latency is the median across invocations; the spread is the min–max over all timed runs. No thermal throttling occurred during these runs (`vcgencmd get_throttled` stayed `0x0`).
120
+
121
+ | File | Inference (median) | Spread (min–max) | Runs | Peak memory |
122
+ |---|---:|---:|---:|---:|
123
+ | `nafnet_fp16.tflite` | 820.9 ms | 812.3–829.7 ms | 150 | 264 MB |
124
+
125
  ## License
126
 
127
  [MIT](https://github.com/megvii-research/NAFNet/blob/main/LICENSE). Upstream: