Z.ai: GLM 5.2 Performance
Compare measured TTFT, throughput, uptime, and route health for Z.ai: GLM 5.2 across TrustedRouter providers using metadata-only production probes.
z-ai/glm-5.2
Measured performance
Continuously sampled p50/p95 time-to-first-token (TTFT), effective throughput, and success rate for Z.ai: GLM 5.2. Effective throughput uses provider-reported output tokens over complete request time. Unsupported route and probe-configuration rows are separated from provider downtime, and no prompt or output content is stored.
| Provider | p50 TTFT | p95 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|
| friendli | 823 ms | 3151 ms | — | 100.00% | — | 6 |
| together | 1094 ms | 2706 ms | 77 tok/s n=1 | 100.00% | — | 2 |
| morph | 1525 ms | 2027 ms | — | 100.00% | — | 3 |
| scaleway | 1592 ms | 4475 ms | — | 100.00% | — | 3 |
| digitalocean | 1666 ms | 1666 ms | 6 tok/s n=1 | 100.00% | — | 1 |
| io-net | 1668 ms | 1668 ms | 122 tok/s n=1 | 100.00% | — | 1 |
| parasail | 1992 ms | 1992 ms | 103 tok/s n=1 | 100.00% | — | 1 |
| engy | 2025 ms | 2426 ms | 36 tok/s n=1 | 100.00% | — | 5 |
| regolo | 2030 ms | 18410 ms | 45 tok/s n=1 | 100.00% | — | 6 |
| inceptron | 2156 ms | 9565 ms | — | 100.00% | — | 5 |
| wandb | 2258 ms | 2258 ms | 150 tok/s n=1 | 100.00% | — | 1 |
| gmi | 3086 ms | 4797 ms | — | 100.00% | — | 7 |
| zai | 3212 ms | 5813 ms | 52 tok/s n=1 | 100.00% | — | 6 |
| atlas-cloud | 3480 ms | 3480 ms | — | 100.00% | — | 1 |
| siliconflow | 4341 ms | 4341 ms | — | 100.00% | — | 1 |
| arcee | 4513 ms | 5339 ms | — | 100.00% | — | 4 |
| baseten | 4944 ms | 11074 ms | — | 100.00% | — | 42 |
| chutes | 7486 ms | 9901 ms | — | 100.00% | — | 2 |
| phala | 10879 ms | 16011 ms | 32 tok/s n=1 | 83.33% | — | 6 |
| alibaba | — | — | 60 tok/s n=1 | — | — | 0 |
| deepinfra | — | — | 37 tok/s n=1 | — | — | 0 |
| fireworks | — | — | 82 tok/s n=1 | — | — | 0 |
| telnyx | — | — | 60 tok/s n=1 | — | — | 0 |
| venice | — | — | 86 tok/s n=1 | — | — | 0 |
Full provider & model leaderboard.
28 routes.
More routes give the auto router more room to fail over around provider 429 and 5xx responses.
Gateway overhead is measured separately.
Public status separates TLS/health overhead from full model latency so slow LLMs do not inflate the router metric.
Metadata rollups.
Status samples store latency, outcome, provider, model, route, cost, and region metadata only.
View public status or inspect provider routes.