Upload benchmarks/streaming.log with huggingface_hub
Browse files- benchmarks/streaming.log +16 -0
benchmarks/streaming.log
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2026-04-09 04:54:01,412 [INFO] Loading faiss with AVX512 support.
|
| 2 |
+
2026-04-09 04:54:01,533 [INFO] Successfully loaded faiss with AVX512 support.
|
| 3 |
+
2026-04-09 04:54:03,332 [INFO] Benchmarking full recompute (50 trials)...
|
| 4 |
+
2026-04-09 04:54:10,527 [INFO] Benchmarking streaming inference (50 trials)...
|
| 5 |
+
|
| 6 |
+
============================================================
|
| 7 |
+
STREAMING INFERENCE BENCHMARK
|
| 8 |
+
============================================================
|
| 9 |
+
Context: 256 units, 10 features
|
| 10 |
+
Device: cuda:0
|
| 11 |
+
------------------------------------------------------------
|
| 12 |
+
Full recompute: 128.72 ± 65.50 ms
|
| 13 |
+
KV-cached: 133.05 ± 78.01 ms
|
| 14 |
+
Speedup: 1.0×
|
| 15 |
+
============================================================
|
| 16 |
+
2026-04-09 04:54:18,907 [INFO] Results saved to outputs/benchmarks/streaming_results.json
|