OpenAI: gpt-oss-120b Performance
TrustedRouter performance signals and provider route posture for OpenAI: gpt-oss-120b.
openai/gpt-oss-120b
Measured performance
Continuously sampled p50/p95 time-to-first-token (TTFT), time-to-first-byte (TTFB), effective throughput, and success rate for OpenAI: gpt-oss-120b. 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 | p50 TTFB | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|---|
| cerebras | 2377 ms | 4512 ms | 2377 ms | — | 100.00% | — | 30 |
| fireworks | 2469 ms | 4874 ms | 2469 ms | — | 100.00% | — | 20 |
| baseten | 2542 ms | 3901 ms | 2542 ms | — | 100.00% | — | 27 |
| nebius | 2543 ms | 3507 ms | 2543 ms | — | 100.00% | — | 8 |
| novita | 2616 ms | 16126 ms | 2616 ms | — | 100.00% | — | 4 |
| crusoe | 2714 ms | 7927 ms | 2714 ms | — | 100.00% | — | 14 |
| parasail | 3108 ms | 5143 ms | 3108 ms | — | 100.00% | — | 9 |
| cloudflare-workers-ai | — | — | — | — | 0.00% | — | 5 |
| tinfoil | — | — | — | — | 0.00% | — | 53 |
Full provider & model leaderboard.
24 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.