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-train)

Backend:       torch
Device:        rtx-5070-ti
Model:         mlp
Mode:          train
Runs:          15, since 2026-10-12
Latest:        91.6k samples/s on 2026-11-02
Best:          95.1k samples/s on 2026-10-21, keras 5d4baf0

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

100% │▁▂▆▆█▃▂ ▂▃  ▆▆▅
 87% │███████▆██▅ ███
 75% │███████████ ███
 62% │███████████ ███
 50% │███████████ ███
 37% │███████████ ███
 25% │███████████▇███
  0% │███████████████
     └───────────────
                   now
TIME↓SAMPLES/SΔ PREVKERAS
2026-11-02T14:01Z91.6k-2.2%df66bff
2026-11-01T10:01Z93.6k+0.6%df66bff
2026-10-31T10:01Z93.0k+306.2%df66bff
2026-10-29T02:01Z22.9k-71.2%7efab4a
2026-10-27T01:01Z79.5k-10.4%bc84c95
2026-10-26T01:01Z88.7k+1.7%cfebb2a
2026-10-25T00:01Z87.2k+8.4%cfebb2a
2026-10-24T00:01Z80.5k-7.5%881611d
2026-10-23T00:01Z87.0k-1.9%e612981
2026-10-22T00:01Z88.7k-6.7%6aae9dc
2026-10-21T00:01Z95.1k+2.4%5d4baf0
2026-10-20T00:01Z92.8k+1.0%1344327
2026-10-19T00:01Z91.9k+5.0%15fdc4e
2026-10-13T13:01Z87.5k+2.7%7bea4c4
2026-10-12T06:01Z85.2k63e75ba