Context:
homelab
Cluster:
k3s
User:
marcos
Forge Rev:
v1.0 (C17)
K8s Rev:
v1.36.5+k3s1
CPU:
18%
MEM:
32%

describe(bench/torch-rtx-5070-ti-mlp-predict)

Backend:       torch
Device:        rtx-5070-ti
Model:         mlp
Mode:          predict
Runs:          15, since 2026-10-12
Latest:        286.0k samples/s on 2026-11-02
Best:          298.6k samples/s on 2026-10-20, keras 1344327

Speed as % of the best run, oldest on the left

100% │▁▃▆█▅▂▃ ▂▂  ▆▂▅
 87% │██████████▂ ███
 75% │███████████ ███
 62% │███████████ ███
 50% │███████████ ███
 37% │███████████▇███
 25% │███████████████
  0% │███████████████
     └───────────────
                   now
TIME↓SAMPLES/SΔ PREVKERAS
2026-11-02T14:01Z286.0k+5.5%df66bff
2026-11-01T10:01Z271.2k-6.4%df66bff
2026-10-31T10:01Z289.6k+170.2%df66bff
2026-10-29T02:01Z107.2k-54.6%7efab4a
2026-10-27T01:01Z236.3k-14.4%bc84c95
2026-10-26T01:01Z275.9k+1.2%cfebb2a
2026-10-25T00:01Z272.8k+3.7%cfebb2a
2026-10-24T00:01Z263.1k-5.3%881611d
2026-10-23T00:01Z277.9k+0.8%e612981
2026-10-22T00:01Z275.7k-3.5%6aae9dc
2026-10-21T00:01Z285.8k-4.3%5d4baf0
2026-10-20T00:01Z298.6k+2.2%1344327
2026-10-19T00:01Z292.1k+5.7%15fdc4e
2026-10-13T13:01Z276.5k+3.4%7bea4c4
2026-10-12T06:01Z267.5k63e75ba