OpenAI: gpt-oss-20b Performance
Try gpt-oss-20b on us. Starter credit covers your first requests, then one key reaches all 13 provider routes with automatic failover.
openai/gpt-oss-20b
Measured performance
Continuously sampled p50/p95 time-to-first-token (TTFT), effective throughput, and success rate for OpenAI: gpt-oss-20b. 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 |
|---|---|---|---|---|---|---|
| parasail | 641 ms | 641 ms | — | 100.00% | — | 1 |
| wandb | 1186 ms | 1186 ms | — | 100.00% | — | 1 |
| akashml | 1251 ms | 2808 ms | — | 100.00% | — | 5 |
| io-net | 1292 ms | 1292 ms | — | 100.00% | — | 1 |
| regolo | 1342 ms | 1381 ms | — | 100.00% | — | 3 |
| databricks | 1679 ms | 3274 ms | — | 100.00% | — | 2 |
| greenference | 2140 ms | 3138 ms | — | 100.00% | — | 5 |
| nvidia-nim | 2952 ms | 2952 ms | — | 100.00% | — | 1 |
| deepinfra | 2976 ms | 2976 ms | — | 100.00% | — | 1 |
| cloudflare-workers-ai | 3160 ms | 3160 ms | — | 100.00% | — | 1 |
Full provider & model leaderboard.
13 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.