Measured on coreweave
8 runs- Avg Tokens / Second
- 53.10
- Decode Tokens / Second
- —
- Avg Time to First Token (ms)
- 1620.00
- Runs Analysed
- 8
- Last Updated
- Oct 13, 2026, 07:00 PM
Throughput distribution
Throughput over time
Generated throughput, including reasoning tokens — so a thinking model's line does not collapse; the table ranks on visible tokens · one cell per model, showing its best-covered provider · models with at least 4 samples in the window first, fastest of those at the top · shared vertical scale, 0 to 160 tok/s (99th percentile) · dashed rule is the model's own mean
Every lane measuring this model
2 measured · live first · a lane is a provider, a catalogue id and a transport| Provider | Lane | Avg Toks/Sec | Min | Max | Avg TTF (ms) | n |
|---|---|---|---|---|---|---|
| coreweave | via OpenRouter | 53.10 | 8.71 | 70.00 | 1620.00 | 8 |
| coreweave | direct | 5.92 | 5.92 | 5.92 | 10400.00 | 1 |
Frequently Asked Questions
How fast is GLM 5.2?
The latest rolling average throughput is 53.10 tokens per second with an average time to first token of 1620.00 ms across 8 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Oct 13, 2026, 07:00 PM.