zai-org

GLM-5.3

Hugging Face

SGLang 0.5.22

NVIDIA B300 x4 · tp4 × pp1
ShareLinkedInX
Speed per user (c=1)
118tok/s
Time to first token
88ms
Peak throughput
5,061tok/s @ 256
Per GPU at peak
1,265tok/s
Cold start
447s
Context tested
8,192tokens

Measured 9 Nov 2026 · Throughput sweep

vLLM 0.32.0

NVIDIA B300 x4 · tp4 × pp1
ShareLinkedInX
Speed per user (c=1)
111.2tok/s
Time to first token
78ms
Peak throughput
5,156tok/s @ 256
Per GPU at peak
1,289tok/s
Cold start
802s
Context tested
8,192tokens

Measured 9 Nov 2026 · Throughput sweep

ModelPerf Pro

What GLM-5.3 costs to run, per token and per watt

  • Cost per million tokens for every model, engine and GPU
  • Energy per million tokens, tokens per watt and GPU power draw
  • Full concurrency curves for all three workload scenarios
  • Unlimited ROI and break-even calculations with PDF export
  • CSV export of the full leaderboard, paid metrics included
  • New models and re-runs included as they are published
$39 / month, billed annually

or $49 month to month

Get ModelPerf ProSign in to buy · $10Compare plans