OpenAI compatible API. Attested gateway. Public status.
Z.ai: GLM 4.7 Performance
TrustedRouter performance signals and provider route posture for Z.ai: GLM 4.7.
1 URLbase_url migration
100smodels and routes
0prompt logs by default
z-ai/glm-4.7
Performance
Measured performance
Continuously sampled p50/p95 time-to-first-token (TTFT), time-to-first-byte (TTFB), throughput, and success rate for Z.ai: GLM 4.7 — unsupported route and probe-configuration rows are separated from provider downtime, and no prompt or output content is stored.
| Provider | p50 TTFT | p95 TTFT | p50 TTFB | Throughput | Uptime | Config excluded | Samples |
|---|---|---|---|---|---|---|---|
| cerebras | 1131 ms | 5029 ms | 1125 ms | — | 40.51% | — | 79 |
| venice | 1357 ms | 3444 ms | 1267 ms | — | 68.18% | — | 22 |
| zai | 1933 ms | 2616 ms | 1930 ms | 42 tok/s | 41.38% | — | 29 |
| phala | 2553 ms | 10618 ms | 2450 ms | — | 64.71% | — | 17 |
| parasail | — | — | — | — | 0.00% | — | 9 |
Full provider & model leaderboard.
Provider diversity
10 routes.
More routes give the auto router more room to fail over around provider 429 and 5xx responses.
Streaming
Gateway overhead is measured separately.
Public status separates TLS/health overhead from full model latency so slow LLMs do not inflate the router metric.
Status
Metadata rollups.
Status samples store latency, outcome, provider, model, route, cost, and region metadata only.
View public status or inspect provider routes.